Novel antagonists acting at the P2Y1 purinergic receptor: Synthesis and conformational analysis using potentiometric and nuclear magnetic resonance titration techniques

Article abstract of DOI:10.1021/jm0104062

The human P2Y₁ receptor is widely distributed in many tissues and has a classical structure of a G protein-coupled receptor. Activated by adenosine-5′-diphosphate (ADP), this receptor is essential for platelet aggregation. In the present paper,

Full text of DOI:10.1021/jm0104062

962
J . Med. Chem. 2002, 45, 962-972
Novel An ta gon ists Actin g a t th e P 2Y1 P u r in er gic Recep tor : Syn th esis a n d
Con for m a tion a l An a lysis Usin g P oten tiom etr ic a n d Nu clea r Ma gn etic
Reson a n ce Titr a tion Tech n iqu es
Pierre Raboisson,† Anthony Baurand,‡ J ean-Pierre Cazenave,‡ Christian Gachet,‡ Myriam Retat,†
Bernard Spiess,*,† and J ean-J acques Bourguignon*,†
Laboratoire de Pharmacochimie de la Communication Cellulaire, UMR 7081 du CNRS, Faculte´ de Pharmacie,
Universite´ Louis Pasteur, 74, Route du Rhin, 67401 Illkirch Cedex, France, and Laboratoire de Biologie et de Pharmacologie
de l’He´mostase et de la Thrombose, Etablissement Franc¸ais du Sang-Alsace, U.311 de l’INSERM, 10, Rue Spielmann, BP 36,
67065 Strasbourg Cedex, France
Received August 28, 2001
The human P2Y1 receptor is widely distributed in many tissues and has a classical structure
of a G protein-coupled receptor. Activated by adenosine-5′-diphosphate (ADP), this receptor is
essential for platelet aggregation. In the present paper, we describe the synthesis of novel
P2Y1 antagonists that could be of interest at least as tools to define the physiological roles of
the P2Y1 receptor, at best as new antithrombotic agents. Thus, we prepared the 2,N6-dimethyl-
2′-deoxyadenosine-3′,5′-bisphosphate derivative, 1e. The biological activity was demonstrated
by the ability of compound 1e to inhibit ADP-induced platelet aggregation, shape change, and
intracellular calcium rise. This compound was a full antagonist at the P2Y1 receptor with a
pA2 value of 7.11 ( 0.11 and was found to be 4-fold more potent than the reference N6-methyl-
2′-deoxyadenosine-3′,5′-bisphosphate (1a , pA2 ) 6.55 ( 0.05), revealing the potency-enhancing
effects of the 2-methyl group. The better activity of 1e as compared to 1a was analyzed using
both potentiometric and nuclear magnetic resonance titration techniques, which highlighted
specific conformational features of this compound. These results clearly indicate the preference
for both compounds for an anti conformation at the N-glycosyl linkage. Furthermore, the
percentage of S conformer of 1e is close to that of 1a , which is nearly 70% at pH ) 2.8 and
increases dramatically when pH increases. From the macroprotonation constants, it can be
noted that compound 1e is significantly more basic than 1a . This is indeed expected for the
N1 adenine nitrogen due to the electron-donating character of the methyl moiety. By considering
the microconstants of the phosphate groups, the higher basicity of P3 and P5 for 1e may be
due to the decrease in the local dielectric constant induced by the substitution of the hydrogen
atom by a more lipophilic methyl group. Thus, it may be suggested that the gain in activity of
1e when compared to the reference compound 1a would result from its gain in basicity rather
than steric and conformational modifications. The synthesis of the first selective radioligand
acting at the P2Y1 receptor ([33P]-N6-methyl-2′-deoxyadenosine-3′,5′-bisphosphate, 17) is also
reported and will be used in the future for efficient screening needed for drug optimization.
In tr od u ction
recently identified2 coupled to Gi and adenylyl cyclase
inhibition.3 The P2Y1 receptor is necessary for ADP to
induce platelet aggregation, since its inhibition in vitro
by selective antagonists totally abolishes ADP-induced
aggregation and calcium mobilization.3,4 This receptor
has been recently shown to be a promising potential
target for new antithrombotic drugs.3 Indeed, P2Y1 null
mice display strong resistance to the thromboembolism
induced by intravenous injection of a mixture of collagen
and adrenaline5,6 or thromboplastin.7 Moreover, the ad-
ministration to mice of the P2Y1 antagonist N6-methyl-
2′-deoxyadenosine-3′,5′-bisphosphate (1a ) resulted in
prolongation of the bleeding time, inhibition of ex vivo
platelet aggregation in response to ADP, and resistance
to thromboembolism induced by collagen and adrenaline
or tissue factor.7,8
Adenine nucleotides interact with P2 receptors, which
are widely distributed in many different cell types
including endothelial cells, smooth muscle, epithelial
cells, lungs, platelets, the pancreas, and central nervous
system and regulate a broad range of physiological
processes.1 These receptors are divided into two families
of membrane-bound purinergic receptors. The G protein-
coupled or metabotropic receptors are termed P2Y, and
the ligand-gated ion channel or ionotropic receptors are
termed P2X. It is well-established that a normal platelet
response to adenosine-5′-diphosphate (ADP) requires
activation of both the P2Y1 receptor, responsible for the
mobilization of ionized Ca2+ from internal stores, through
a Gq phospholipase C pathway and the P2Y12 receptor
* To whom correspondence should be addressed. B.S. Tel: +33 390
24 42 23. Fax: 33 390 24 43 10. E-mail: spiess@aspirine.u-strasbg.fr.
or J .J .B. Tel: +33 390 24 43 10. Fax: 33 390 24 43 10. E-mail:
jjb@pharma.u-strasbg.fr.
Thus, selective P2Y1 receptor antagonists may have
potential therapeutic use as antithrombotic agents. On
the other hand, the exact physiological role of this
receptor is largely unknown, and few pharmacological
† Universite´ Louis Pasteur.
‡ Etablissement Franc¸ais du Sang-Alsace.
10.1021/jm0104062 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/12/2002

Novel Antagonists Acting at the Purinergic Receptor
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4 963
the preparation of [33P]-N6-methyl-2′-deoxyadenosine-
3′,5′-bisphosphate (17) for binding experiments. Com-
bining pharmacological data with potentiometric and
nuclear magnetic resonance (NMR) analyses allowed us
to highlight some conformational and electronic require-
ments of the ribose bisphosphate moiety as well as of
the adenine ring of these adenine derivatives.
Ch em ica l Syn th esis
As shown in Scheme 1, 2,N6-dimethyl-2′-deoxyaden-
osine (9) was prepared by ring fission of the 1,N6-etheno
derivative 4 and subsequent unmasking of the etheno
moiety.17-23
2′-Deoxyadenosine (3) was converted into the 1,N6-
etheno derivative 4 with chloroacetaldehyde according
to a previously reported procedure.17 The presence of
the etheno group rendered position 2 of the adenine ring
sensitive to nucleophilic attack. Thus, treatment of 4
with aqueous sodium hydroxide18 afforded the imidazole
derivative 5 in 86% yield. Cyclocondensation of 5 with
trimethylorthoacetate18 led to the 2-methyl derivative
6, which, after treatment with ammonium persulfate
phosphate buffer at pH 7.2,19 afforded the resulting
adenine (7) in good yield. The reaction of the latter with
1,2-bis[(dimethylamino)methylene]hydrazine20 in re-
fluxing pyridine21 afforded the 6-(1,2,4-triazol-4-yl) de-
rivative 8 in 50% yield. Nucleophilic displacement of
this triazole moiety in 8 with methylamine21 led to
the corresponding 2,N6-dimethyl derivative 9. Finally,
phosphorylation of 9 using tetrabenzylpyrophosphate
(TBPP),22,23 followed by catalytic hydrogenation, af-
forded the expected 2-methyl derivative 1e in 76%
yield.
The radioligand 17 was synthesized as shown in
Scheme 2.24 The N6-methyl-2′-deoxyadenosine was first
reacted with 4,4′-dimethoxytrityl chloride24 to give the
5′-protected derivative 13, followed by phosphorylation
with TBPP to afford compound 14 in 93% yield.22,23
Deprotection of the DMT group was achieved by treat-
ment with 90% formic acid for 10 min.25 The catalytic
hydrogenation of compound 15 led to the monophos-
phorylated compound 16 in 83% yield. Finally, its
phosphorylation with 33P was achieved by enzyme-
catalyzed trans phosphorylation using polynucleotidase
kinase at the Du Pont NEN Laboratory (Le Blanc
Mesnil, France).
F igu r e 1. Structures of nucleotides analogues acting on the
P2Y1 receptor.
tools are available today. Thus, the development of
potent and selective P2Y1 receptor antagonists is of a
critical need.
The recently reported adenine nucleotide bisphos-
phate 1a (Figure 1)9 has a KB value at the human P2Y1
receptor of 100 nM9 and was inactive at other P2Y
receptors subtypes10,11 including P2X2 and P2X4 recep-
tors.12 However, the selectivity profile of 1a for the P2Y1
receptor was not optimized as it displayed significant
affinity at P2X1 and P2X3 receptors (IC50 ) 1.15 and
12.9 µM, respectively).12
More recently, 2′-deoxyadenosine bisphosphate de-
rivatives, structurally related to 1a and presenting
various structural modifications at positions 2 and 6 of
the adenine ring, have been synthesized with the goal
of developing more potent and selective P2Y1 antago-
nists.13
Substitutions at position 2 of adenosine-5′-triphos-
phate nucleotides (2; Figure 1) are known to be tolerated
and in some cases are favorable for P2Y receptor
agonists.14 Thus, it is not surprising that the introduc-
tion of 2-methylthio and 2-chloro substituents in 2′-
deoxyadenosine-3′,5′-bisphosphate led to partial ago-
nists nearly 10 times more potent than the 2-unsubsti-
tuted compound.13 Moreover, the introduction of the
N6-methyl group in these 2-methylthio and 2-chloro
derivatives (compound 1c,d ; Figure 1) decreased agonist
efficacy, affording pure P2Y1 receptor antagonists.15 On
the other hand, major synthetic modifications of the
ribose moiety have been carried out to increase biologi-
cal stability and selectivity for the receptors.15,16
However, to our knowledge, the effects of the intro-
duction of an alkyl group at position 2 have not been
described. To better understand the structure-activity
relationships (SAR) of 1a structurally related com-
pounds acting to the P2Y1 receptors, the 2-methyl
derivative 1e has been synthesized and tested in vitro
in comparison to 1a .
Biologica l Activity
P la telet Aggr ega tion . The new derivative 1e pre-
pared in the present study was first tested as an
antagonist in a platelet aggregation assay. The addition
of 1e (20 µM) to washed human platelets took place 30
s before ADP (5 µM) inhibited platelet aggregation and
shape change (Figure 2A), while 1e alone did not induce
shape change or aggregation even at high concentra-
tions (up to 100 µM). The nature of inhibition was
determined by generating a series of concentration-
response curves for ADP in the presence of different
concentrations of 1e. This compound caused a parallel
shift to the right of the concentration-response curve,
but high concentrations of ADP could completely over-
ride high concentrations of 1e (Figure 2B). Schild
analysis of the inhibition gave a pA2 value of 7.11 ( 0.11
(n ) 3) and a slope of 0.66 (Figure 2B inset), which could
Its activity at P2Y1 receptors was determined by
measuring their capacity to inhibit platelet aggregation
in vitro. Furthermore, for further works dealing with
drug optimization, the critical need of a radiolabeled
ligand of the P2Y1 receptor encouraged us to develop

964 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4
Sch em e 1. Synthesis of 1ea
Raboisson et al.
a
Reagents: (a) CHOCH2Cl; (b) NaOH; (c) CH3C(OMe)3; (d) (NH4)2S2O8; (e) (CH3)2NsNdNsN(CH3)2; (f) NH2CH3; (g) t-BuOK, TBPP;
(h) H2, Pd/C.
Sch em e 2. Synthesis of 17a
of 1e, in the presence or absence of 2 mM external Ca2+
(Figure 3A). Conversely, 100 µM of compound 1e had
no influence on basal levels of cyclic adenosine-3′,5′-
monophosphate (cAMP) in human platelets (Figure 3B)
or on the cAMP levels induced by 10 µM prostaglandin
E1 (PGE1). The ability of ADP to inhibit PGE1-stimu-
lated cAMP accumulation was likewise not affected by
compound 1e in human platelets (Figure 3B), whereas
AR-C66096,26 a selective P2Y12 receptor antagonist,
totally reversed the inhibitory effect of ADP.
P oten tiom etr ic Stu d ies a n d NMR
Deter m in a tion s
Ma cr oscop ic a n d Micr oscop ic P r oton a tion Con -
sta n ts. The studied compounds carry, in addition to a
protonable nitrogen on adenine, two phosphate groups,
and each group is able to bind only one proton for pH
values ranging from 10 to 2.5. Thus, in the latter pH
range, three macroscopic overall protonation constants
a
Reagents: (a) DMTCl, DMAP; (b) t-BuOK, TBPP; (c) HCOOH;
(d) H2, Pd/C; (e) polynucleotidase kinase.
be explained by the fact that we observed an integrated
aggregation process involving the activation of two
receptors (P2Y1 and P2Y12) and their transduction
machinery. Compound 1e was considerably more potent
(about 4-fold) as an antagonist as compared to 1a (pA2
) 6.55 ( 0.05).8 As seen with compound 1a , no agonist
activity was observed.8 It is interesting to note that the
monophosphate derivative 16 was found inactive. This
result emphasizes the critical role played by the phos-
phate in position 5′ of this series of antagonists, in
agreement with earlier literature data.9
Ca lciu m Mobiliza tion a n d In h ibition of Ad en y-
lyl Cycla se. In platelets, ADP induced simultaneous
mobilization of intracellular Ca2+ stores and inhibition
of adenylyl cyclase, through activation of the P2Y1 and
P2Y12 receptors, respectively. We have verified that
compound 1e acts selectively on the P2Y1 receptor. The
intracellular Ca2+ rise induced in washed human plate-
lets by 5 µM ADP could be totally inhibited by 100 µM
ây (with y ) 1-3) quantify the protonation process
ây
according to the following equilibrium: L4- + H+
{
}
y
HyL(4-y)-. When defined step by step, the protona-
tion process may be quantified by Ky, characterizing
Ky
the equilibrium Hy-1L(5-y)- + H+ } HyL(4-y)-. It can
{\
be noted that log Ky corresponds to the usual pKa
value.
These constants, easily determined by potentiomet-
ric27 or NMR titration methods,28 cannot be attributed
to a given protonation site since most of them are
less than two log units apart; therefore, a given mac-
roscopic protonation step involves two different basic
sites. An inframolecular approach, which aims at de-
fining the intrinsic acid-base properties of each indi-
vidual functional group, thus requires the resolution of
a more detailed protonation scheme, which, in the
case of compounds 1e and 1a , is depicted in Figure 4.

Novel Antagonists Acting at the Purinergic Receptor
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4 965
F igu r e 3. Effects of the compound 1e on P2Y1 and P2Y12
receptor transduction pathways. (A) Added 30 s before ADP,
100 µM compound 1e, totally abolished the [Ca2+]i rise induced
by 5 µM ADP in washed human platelets in the presence of 2
mM external Ca2+. (B) In the presence of 100 µM compound
1e (hatched bars), ADP was still able to reduce PGE1-
stimulated cAMP accumulation in washed human platelets.
Data are from one experiment representative of three inde-
pendent experiments giving identical results.
F igu r e 2. Effect of the compound 1e on ADP-induced platelet
aggregation and shape change. Platelet aggregation was
measured as described.58 The ordinate represents apparent
changes in optical density (light transmission) due to the light
scattering by the platelets. (A) Aggregation and shape change
in response to 5 µM ADP (control) were inhibited by 20 µM of
the compound 1e. Traces are from one experiment representa-
tive of three independent experiments giving identical results.
(B) Aggregation was induced by increasing concentrations of
ADP alone or in the presence of increasing concentrations of
the compound 1e added 30 s before ADP. (Inset) Schild
regression analysis of the data shown in panel B. Curves
represent the mean of three independent experiments and give
a pA2 value of 7.11 ( 0.11 and a Schild slope of 0.66. Bars
show the standard error of mean.
In that figure, the two first protonation steps, which
refer to the two phosphate groups, have to be described
by four microspecies and four related microprotonation
constants. As will be shown later, the third equivalent
of added protons mainly binds to an adenine nitrogen,
so that K3 satisfactorily defines the last protonation
step.
F igu r e 4. Microscopic protonation scheme for the studied
adenosine-3′,5′-bisphosphates.
shown,35 the individual protonation fractions fi,p may
be expressed as a function of the macro- and micropro-
tonation constants
â2[H+]2 + ki[H+]
â2[H+]2 + â1[H+] + 1
31P NMR has proven to be a good probe to study
individual protonation,29-40 provided that the observed
chemical shifts for the phosphorus resonances δiobs
mainly depend on the electronic effects accompanying
the variations in the protonation states. In that case,
the protonated fraction fi,p of a phosphate group in
position i on compounds 1e or 1a can be calculated by
the eq 1:
fi,p
)
(2)
where ki refers to k3 or k5 of the microequilibria
displayed in Figure 4. k35 and k53 can be further
calculated knowing that K1K2 ) k3‚k35 ) k5‚k53. Equa-
tion 2 is solved by nonlinear regression introducing the
macroprotonation constants obtained by the NMR ex-
periments to give the microprotonation constants.
δoi bs - δi,d
δi,p - δi,d
fi,p
)
(1)
Resu lts a n d Discu ssion
A plot of the chemical shift vs pH for the phosphorus
nuclei of 1a is shown in Figure 5a. The curves for 1e
have the same general shape as those of Figure 5, only
they are shifted in their steepest part to higher pH
where δi,p and δi,d correspond, respectively, to the
chemical shifts of the protonated and deprotonated
fractions of the phosphates in position i. As previously

966 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4
Raboisson et al.
F igu r e 5. Chemical shifts (δ) from 31P NMR titrations for 1a
(a) and the corresponding protonation fraction curves fi,p (b)
as a function of pH in KCl 0.2 M at 37 °C (2H2O). The least-
squares fit of fi,p vs pH according to type 2 equations is shown
in the solid line in panel b.
F igu r e 6. Relative concentrations of the protonated microspe-
cies of compounds 1a (a) and 1e (b) in KCl 0.1 M at 37 °C,
plotted against pH.
Ta ble 1.a
cant basicity increase of the three functional groups for
1e with respect to 1a . By considering the interactivity
parameter previously defined,35 it can be observed that
even though they are weak, the interactions between
phosphates P3 and P5 are higher for 1e than for 1a .
Figure 6 shows the distribution of the various micro-
protonated species as a function of pH, providing a
direct observation of the protonation state of the func-
tional groups.
1H NMR titration experiments were performed si-
multaneously to the 31P NMR titrations in order to gain
complementary structural information. The titration
curves of Figure 7a,b correspond to the proton reso-
nances of, respectively, compounds 1a and 1e that are
satisfactorily resolved over the entire studied pH range.
Upon protonation, i.e., from pH 11.5 to pH 2.0, it can
be seen that most of the protons move significantly
downfield (H2, CH311, H2′, H3′, and H5′ for 1a and
CH32, CH311, H2′, H3′, and H5′ for 1e), some others
are only slightly affected (H1′ and H4′), and one of them
(H8) undergoes chemical shift variations first to higher
fields and then in the opposite direction. It must be
recalled that in the most general case, the binding of a
proton to a basic site leads to an electron density
decrease and thus, via a through bond effect, to a shift
of the proton resonances to lower fields. The opposite
trend, called “wrongway shift”, which has already been
ligand
y
log ây
log Ky
i
log ki
ii′
log kii′
1a
1
2
3
1
2
3
6.68
11.96
15.87
7.08
13.02
17.69
6.68
5.28
3.91
7.08
5.94
4.67
3
5
6.40
6.48
35
53
5.55
5.47
1e
3
5
6.76
6.86
35
53
6.26
6.16
a
Logarithms of the overall (log ây) and stepwise macroproto-
nation constants (log Ky) and microprotonation constants according
to Figure 5 for compounds 1a and 1e. log ki and log kii′ represent
a general designation for, respectively, the logarithms of the first
and second stepwise microprotonation constants. i and ii′ allow
the location of the protons on the phosphates of the studied
compounds. The calculated interactivity parameter ∆log k3-5
)
log k3 - log k53 ) log k5 - log k35 ) 0.93 for 1a and 0.60 for 1e. It
can be noted that log Ky also corresponds to the classical pKa values
that refer to a proton dissociation process.
values by about 0.4 pH units. The monophasic shape of
these curves indicates only weak interactions between
phosphates P3 and P5, each phosphate group having a
large conformational freedom. The protonation fraction
curves of Figure 5b show, in addition, that both phos-
phates display about the same basicity, with P5 only
slightly more basic than P3. The macro- and micropro-
tonation constants for 1e and 1a listed in Table 1 do
not only confirm the equivalent basicity of the phos-
phates for both compounds but also indicate a signifi-

Novel Antagonists Acting at the Purinergic Receptor
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4 967
protonation occurring at the N1 adenine nitrogen since
the electron density changes at this site can be easily
transmitted through the π system of the aromatic ring
to more distant places. This has been described earlier
about many adenine derivatives.43,44 Also, expected
downfield shifts of about 0.16 ppm arise for H3′ and H5′,
centered around pH 6.5, which corresponds to the
inflection point of P3 and P5. This clearly results from
the electron-withdrawing effect of both phosphates on
H3′ and H5′. Although less important, the same effect
operates on the phosphate remote H2′ and H4′ protons.
Finally, the wrongway shift of H8 attests the electro-
static field effect experienced by this proton when
protonation of the phosphate groups occurs. This clearly
shows the preference for both compounds for an anti
conformation at the N-glycosyl bond. A recently pub-
lished work on C2- or C8-modified nucleotides47 also
concluded that there was a preferential anti conforma-
tion for the 2-substituted derivatives and a syn confor-
mation for most of the 8-substituted nucleotides due to
steric hindrance. These observations further lead to the
conclusions that anti type nucleotides are needed to
tightly fit the binding cavity of the P2Y1 receptor.
Regarding the anti- or syn-N-glycoside conformational
requirements, our results are fully in line with those of
literature reports.16,47
However, in addition to the orientation at the N-gly-
cosyl bond, the two-state conformational equilibrium of
the pentofuranosyl moiety of the compounds under
study must also be considered in order to further
delineate the steric and conformational demands for
optimal binding in the P2Y1 receptor cleft. NMR studies
on nucleosides and nucleotides showed that north (N)
(C3′-endo-C2′-exo) and south (S) (C2′-endo-C3′-exo) con-
formations are the dominant forms in the two-state
pseudorotational equilibrium (N T S) in solution.48-51
The pseudorotational parameters and percentage popu-
lations N and S can be accurately calculated from the
proton-proton coupling constants for the ribose moiety
by using a generalized Karplus equation.52,53 However,
in a simpler approach,54,55 the percentage of N and S
3
3
type conformers may be obtained from J 1′,2′ and J 3′,4′
by the following equation: S(%) ) [3J 1′,2′/(3J 1′,2′ + 3J 3′,4′)]
× 100, where the sum of 3J 1′,2′ and 3J 3′,4′ should be nearly
10 Hz. The H-H coupling constants of compound 1e at
3
3
3
pH 2.8 were as follows: J 1′,2′ ) J 1′,2′′ ) 6.8 Hz, J 2′,2′′
3
3
3
) 14.1 Hz, J 2′,3′ ) 5.6 Hz, J 2′′,3′ ) 2.7 Hz, J 3′,4′ ) 2.8
Hz, and J 4′,5′ ) 3.6 Hz. Thus, the percentage of the S
3
conformer of 1e is calculated to be nearly 70%. The
coupling constants of 1a are close to those of 1e, so that
about the same percentage of S conformer can be
calculated for both compounds. These results are in good
agreement with those published for 3′-AMP and 3′-2′-
deoxyAMP,56 which take up, at 278 K, respectively, a
74 and 76% S type conformation.
F igu r e 7. Chemical shifts (δ) from a 1H NMR titration for
1a (a) and for 1e (b) as a function of pH in 0.2 M KCl at 37 °C
(2H2O).
observed in nucleotides,41-45 inositol-phosphates,29,46
and natural compounds such as adenophostin A,30
occurs when a highly negatively charged phosphate
group approaches a hydrogen atom. This effect, elec-
trostatic in origin, operates through the field29,43,44 and
affects the chemical shifts of the hydrogen atoms in the
neighborhood of the phosphate groups, thus providing
valuable structural and conformational information.
Interestingly, by increasing pH, the coupling con-
3
stants slightly vary. Thus, for instance, 3J 1′,2′ and J 1′,2′′
differentiate from pH 4.5 to pH 9.5: the former in-
creases by ca. 1 Hz, and the latter decreases by ca. 0.5
Hz. For the highest pH values, 3J 1′,2′ reaches 8 Hz, which
corresponds to the limiting values for the S conformer.50
In other words, the percentage of S conformer increases
when pH increases. An inspection of a simple molecular
model shows that the S conformation of the ribose
From the curves in Figure 7a,b, it can be seen that
H2 or CH32, H8, and CH311 are sensitive to the

968 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4
Raboisson et al.
moiety ensures a minimal repulsion between the P3 and
the P5 phosphates groups and that the repulsion
increase expected with the deprotonation of these
phosphates drives the sugar pseudorotational equilib-
rium to the S conformer.
acter of the substituents accounts for the observed
biological results, other effects such as changes in the
H-bonding pattern of the adenyl moiety cannot be
disregarded.
In summary, we have synthesized and evaluated the
2,N6-dimethyl-2′-deoxyadenosine-3′,5′-bisphosphate de-
rivative, 1e. The chosen chemical pathway allowed us
to explore in a very expeditive way different substitu-
tions at position 2 by means of the corresponding ortho
ester and by using the amino-imidazol as a common
intermediate leading to possible topological exploration
at this position. The 2-methyl derivative (1e) was tested
as an antagonist of platelet aggregation. At the intra-
cellular level, as expected for a P2Y1 receptor antago-
nist, compound 1e totally inhibited the intracellular
Ca2+ mobilization induced by ADP but had no effect on
its inhibition of adenylyl cyclase. Thus, compound 1e
was significantly more potent (about 4-fold) as an
antagonist as compared to 1a (pA2 ) 6.55 ( 0.05),8
revealing the potency-enhancing effects of the 2-methyl
group. As seen with compound 1a , no agonist activity
was observed.8
To better understand the SAR of 1a structurally
related compounds acting on the P2Y1 receptor, we have
used an integrated approach combining NMR and
potentiometric titration experiments, which led to the
following observations: (i) both compounds 1a and 1e
possess an anti conformation at the N-glycosyl linkage,
and the percentage of S conformer of 1e is close to that
of 1a , nearly 70% at pH 2.8 and increasing dramatically
when pH increases; (ii) from the macroprotonation
constants, it can be noted that compound 1e is signifi-
cantly more basic than 1a at the N1 adenine nitrogen,
due to the electron-donating character of the methyl
moiety; (iii) by considering the microconstants of the
phosphate groups, the higher basicity of P3 and P5 for
1e may be due to the decrease in the local dielectric
constant induced by the substitution of a hydrogen atom
by a more lipophilic methyl group. Thus, it may be
suggested that the gain in receptor affinity of 1e with
regard to 1a also reflects its gain in basicity rather than
steric and conformational modifications.
In a recently published paper,16 molecular modeling
studies were carried out in order to analyze the sugar
conformational requirements for bisphosphate ligands,
among them 1a , for optimal binding to the human P2Y1
receptor. From these studies, the N conformation ap-
peared to be essential to maximize the electrostatic
interactions between the negatively charged phosphate
groups and the positively charged amino acids (Arg128-
(TM3), Lys280(TM6), and Arg310(TM7)) present in the
receptor binding cleft. Because only one pseudorota-
tional form is expected to be present at the binding site
of the receptor, the formation of the ligand-receptor
complex would require a conformational flip from S to
N of high energetic cost, which is likely to be detrimental
to optimal binding. If undoubtedly in our experimental
conditions the S conformer largely predominates, it is
well-known that in solution, various steric and stereo-
electronic effects of the sugar skeleton and the nucleo-
base dictate the N T S pseudorotational equilibrium.
It is, therefore, also possible that in biological conditions,
factors such as the dielectric constant of the medium
and the presence in the neighboring environment of
these polyfunctional ligands of inorganic (Ca2+, Mg2+
K
,
+ ...) or organic (spermine, spermidine ...) cations may
greatly influence the conformational properties of the
studied adenosine-3′,5′-bisphosphates.
In light of the physicochemical properties of both
compounds, what could be the reasons for the higher
affinity of 1e with regard to 1a for the P2Y1 receptor?
From the macroprotonation constants, it can be noted
that compound 1e is significantly more basic than 1a .
This is indeed expected for the N1 adenine nitrogen due
to the electron-donating character of the methyl moiety,
which could maximize the hydrogen-bonding interaction
between the N6 amine of the adenine moiety and the
Gln307(TM7) and/or Ser314(TM7) present within the
binding site. However, by considering the microcon-
stants of the phosphate groups, the higher basicity of
P3 and P5 for 1e is much more surprising since the
inductive electronic effect previously evoked can no
longer apply for these remote phosphate groups. Our
hypothesis is that the observed basicity increase at least
partly stems from the decrease in the local dielectric
constant induced by the substitution of a hydrogen atom
by a more lipophilic methyl group. Because it has been
shown that the affinity of phosphate groups for various
polyamines increases by increasing their basicity, it may
be suggested that the gain in the receptor affinity of 1e
with regard to 1a partially reflects its gain in basicity.
We also hypothesize that the enhanced affinity reported
for the 2-methylthio 1c or chloro 1d derivatives16 may
be attributed to the same effect. For the latter, for
instance, the withdrawing electron effect of the chlorine
atom may lower the basicity of the adenine nitrogen but
because it is as lipophilic as the methyl group, it may
increase the basicity of the phosphate groups. Moreover,
it was not surprising that the more hydrophilic 2-NH2
derivative was less active as the 2-chloro, 2-methyl, and
2-thiomethyl ones.16 Even though the lipophilic char-
The synthesis of the radiolabeled ligand of the P2Y1
receptor 17 has also been reported. As previously
described,8 this compound is now readily available for
further binding studies.
Exp er im en ta l Section
Ch em ica l Syn th esis. Reagents used for the synthesis were
purchased from Sigma-Aldrich (Isle d’Abeau Chesnes, France)
and Lancaster (Bischheim-Strasbourg, France). With the
exception of tetrahydrofuran (THF) and Et2O, all solvents were
obtained from commercial suppliers and used without further
purification. These two solvents were freshly distilled from
sodium benzophenone ketyl. Flash chromatography was per-
formed on Geduran silica gel Si 60 (40-63 µm, Merck). Thin-
layer chromatography was carried out using plates of silica
gel 60 F254 (Merck). The spots were visualized either under
UV light (λ ) 254 nm) or by spraying with molybdate reagent
-
(H2O/concentrated H2SO4/(NH4)6Mo7O24‚4H2O/(NH4)2 Ce(SO4)4‚
2H2O, 90/10/25/1, v/v/w/w) and charring at 140 °C for a few
minutes. All chemical yields are unoptimized and generally
represent the result of a single experiment.
1H NMR were recorded on a Bruker AC 200 (200 MHz) or
a Bruker DPX 300 (300 MHz) spectrophotometer at room
temperature. Chemical shifts are given in parts per million
(δ), coupling constants (J ) are in hertz (Hz), and signals are

Novel Antagonists Acting at the Purinergic Receptor
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4 969
designated as follows: s, singlet; d, doublet; t, triplet; q,
quadruplet; quint, quintuplet; m, multiplet; br s, broad singlet;
etc. The mass spectra were obtained on a Mariner API-TOF.
30 min. The residue was dissolved in anhydrous pyridine (5
mL). The resulting solution was cooled to 0 °C, and thenm
TMSCl (235 µL, 1.85 mmol) was added. The mixture was
heated at 100 °C under argon for 24 h and cooled to 0 °C, and
then, additional TMSCl (94 µL, 0.74 mmol) was added. After
the solution was stirred for 15 min at 20 °C, the solvent was
evaporated to dryness and the residue was dissolved in ice-
cold CH2Cl2 (80 mL). This solution was successively washed
with a cold mixture of brine (40 mL) and saturated NaHCO3
(20 mL), brine (40 mL) and 1 N HCl (15 mL), and finally with
water (20 mL). The organic layer was dried (Na2SO4) and
concentrated under reduced pressure. The residue was dis-
solved in absolute MeOH (15 mL), and the resulting solution
was stirred at room temperature overnight. The solvent was
evaporated, and a mixture of CH2Cl2 (8 mL) and MeOH (1 mL)
was added. The precipitate was filtered, washed with CH2Cl2,
and then dried in vacuo to give 8 (149 mg, 50%) as a white
Melting points were determined with
a Mettler FP62
apparatus and are uncorrected. Elemental analyses were
performed by the CNRS department of microanalysis (CNRS,
Vernaison, France) and are indicated only by the elemental
symbols within (0.4% of the theoretical values unless other-
wise noted.
2-Meth yl-2′-d eoxya d en osin e (7). To a solution of 3 (835
mg, 3.10 mmol) in H2O (30 mL) and THF (30 mL) was added
a 50% solution of chloroacetaldehyde in water (6,50 mL, 51.2
mmol) at 20 °C with stirring, and the pH was maintained at
4-5 with saturated NaHCO3. After the solution was stirred
for 4 days at 20 °C, the solvent was evaporated to dryness and
the residue was purified on silica gel (CH2Cl2/AcOEt/EtOH,
50:40:10) to give a white solid, which was recrystallized from
1
1
solid. H NMR (200 MHz, DMSO-d6): δ 2.34-2.46 (m, 1H, 2′-
EtOH and Et2O to give 4 (735 mg, 83%) as a white solid. H
Ha), 2.72-2.85 (m, 1H, 2′-Hb), 2.77 (s, 3H, CH3), 3.53-3.71
(m, 2H, 5′-H), 3.90-3.97 (m, 1H, 4′-H), 4.43-4.51 (m, 1H, 3′-
H), 5.07 (s, 1H, 3′-OH), 5.42 (s, 1H, 5′-OH), 6.52 (t, 1H, J )
6.6 Hz, 1′-H), 8.93 (s, 1H, 8-H), 9.62 (s, 2H, triazole); m/z 318
(M + H)+.
NMR (200 MHz, CDCl3): δ 2.33-2.44 (m, 1H, 2′-Ha), 2.53-
2.83 (m, 1H, 2′-Hb), 3.51-3.71 (m, 2H, 5′-H), 3.88-3.95 (m,
1H, 4′-H), 4.41-4.48 (m, 1H, 3′-H), 5.02 (t, 1H, J ) 5.7 Hz,
3′-OH), 5.41 (d, 1H, J ) 3.90 Hz, 5′-OH), 6.47-6.53 (m, 1H,
1′-H), 7.58 (d, 1H, J ) 0.84 Hz, etheno Ha), 8.10 (d, 1H, J )
0.84 Hz, etheno Hb), 8.56 (s, 1H, 8-H), 9.31 (s, 1H, 2-H); m/z
276 (M + H)+.
A solution of 8 (150 mg, 0.472 mmol) and methylamine (1.0
M in THF, 3 mL) in absolute ethanol (3 mL) was heated at
100 °C in a sealed tube for 12 h. After the solvent was
evaporated, the residue was chromatographed on silica (AcOEt/
CH2Cl2/EtOH, 4:5:1) and then recrystallized from ethanol to
give 9 (118 mg, 90%) as a white solid; mp 150 °C. 1H NMR
(200 MHz, CDCl3): δ 2.26-2.36 (m, 1H, 2′-Ha), 2.61 (s, 3H,
CH3), 3.10-3.25 (m, 1H, 2′-Hb), 3.24 (m, 3H, CH3), 3.78-4.09
(m, 2H, 5′-H), 4.28-4.31 (m, 1H, 4′-H), 4.82-4.89 (m, 1H, 3′-
H), 5.98 (s, 1H, NH), 6.29-6.37 (m, 1H, 1′-H), 7.74 (s, 1H, 3′-
OH), 8.25 (s, 2H, 8-H, 5′-OH); m/z 280 (M + H)+. Anal.
(C12H17N5O3‚0.5 H2O) C, H, N.
2,N6-Dim et h yl-2′-d eoxya d en osin e-3′,5′-b isp h osp h a t e
(1e). Potassium tert-butoxide (1.0 M in THF, 1.10 mL) was
slowly added, at -40 °C, to a stirred solution of 9 (140 mg,
0.50 mmol) in anhydrous THF (10 mL). After 5 min, tetra-
benzyl pyrophosphate (592 mg, 1.10 mmol) was added and
stirring was continued for 15 min at -40 °C. The reaction
mixture was allowed to warm to 0 °C and was then quenched
with acetic acid (50 µL). The mixture was diluted with ethyl
acetate, washed with water, dried (Na2SO4), and concentrated
to dryness under reduced pressure. Chromatography on silica
(AcOEt/CH2Cl2/EtOH, 40:50:10) afforded the desired product
10 (264 mg, 66%) as a colorless syrup. 1H NMR (200 MHz,
CDCl3): δ 2.40-2.52 (m, 1H, 2′-Ha), 2.57-2.77 (m, 1H, 2′-Hb),
2.60 (s, 3H, CH3), 3.23 (d, 3H, J ) 4.0, NCH3), 4.08-4.29 (m,
3H, 5′-H, 4′-H), 4.98-5.12 (m, 9H, 3′-H, 4CH2), 5.72 (m, 1H,
NH), 6.28-6.36 (m, 1H, 1′-H), 7.29-7.38 (m, 20H, 4Ph), 8.23
(s, 1H, 8-H); m/z 800 (M + H)+.
A mixture of 10 (120 mg, 0.15 mmol) and 10% Pd/C (100
mg) in absolute methanol (20 mL) was shaken in a hydrogena-
tion apparatus under atmosphere pressure at room tempera-
ture for 48 h. The catalyst was removed by filtration and
washed with water, and the filtrate was concentrated to
dryness. Recrystallization from methanol yielded compound
1e (50 mg, 76%) as a colorless solid; mp 146 °C. 1H NMR (200
MHz, DMSO-d6): δ 2.44 (s, 3H, CH3), 2.50-2.65 (m, 1H, 2′-
Ha), 2.80-2.90 (m, 1H, 2′-Hb), 2.94 (m, 3H, CH3), 3.95-4.12
(m, 2H, 5′-H), 4.22-4.35 (m, 1H, 4′-H), 4.93-5.04 (m, 1H, 3′-
H), 6.36 (t, 1H, J ) 4.5, 1′-H), 7.72 (s, 1H, NH), 8.25 (s, 1H,
8-H). 31P NMR (300 MHz, D2O): δ 2.74 (s, 1P, 3-P), 3.10 (s,
1P, 5-P). Anal. (C12H19N5O9P2‚0.2H2O) C, H, N.
N6-Meth yl-2′-d eoxya d en osin e-3′-p h osp h a te (16). Com-
pound 1257 (400 mg, 1.51 mmol) was dissolved in pyridine (10
mL) and then evaporated to dryness. This operation was
repeated twice, and then, the residue was suspended in dry
pyridine (10 mL) under an argon atmosphere. 4,4′-Dimethoxy-
trityl chloride (511 mg, 1.51 mmol), triethylamine (210 µL, 1.51
mmol), and 4-(dimethylamino)pyridine (10 mg, 0.0819 mmol)
were then added, and the resulting solution was stirred for 6
h at room temperature. After the solvents were evaporated,
A solution of 4 (1.00 g, 3.63 mmol) in 0.5 N NaOH (50 mL)
was stirred under reflux for 2 min and then cooled to room
temperature. The solution was neutralized with 1 N HCl. The
mixture was evaporated to dryness. The residue was triturated
in EtOH (150 mL) and then filtered and washed with EtOH.
The combined filtrates were evaporated to dryness, and the
residue was redissolved in EtOH (7 mL). To this solution was
added 1 mL of Et2O. The mixture was stirred at room
temperature for 0.5 h. The precipitate was filtered, washed
with Et2O, and then recrystallized from EtOH and Et2O to
1
yield 5 (830 mg, 86%) as a white powder. H NMR (200 MHz,
dimethyl sulfoxide (DMSO)-d6 + D2O): δ 2.11-2.19 (m, 1H,
2′-Ha), 2.41-2.51 (m, 1H, 2′-Hb), 3.50-3.55 (m, 2H, 5′-H),
3.78-3.82 (m, 1H, 4′-H), 4.31-4.35 (m, 1H, 3′-H), 5.92-5.98
(m, 1H, 1′-H), 6.87 (s, 2H, etheno H), 7.40 (s, 1H, 2-H); m/z
266 (M + H)+.
A mixture of 5 (300 mg, 1.13 mmol), trimethylorthoacetate
(2.16 mL, 17 mmol), trifluoroacetic acid (200 µL, 2.60 mmol),
and anhydrous dimethylformamide was heated at 100 °C
overnight under an argon atmosphere and then concentrated
to dryness. The residue was purified on silica gel (AcOH/
MeOH/CH2Cl2, 5:20:75) to give a white solid, which was
recrystallized from EtOH and Et2O to give 6 (265 mg, 81%)
1
as a white solid. H NMR (200 MHz, DMSO-d6): δ 2.31-2.42
(m, 1H, 2′-Ha), 2.68-2.82 (m, 1H, 2′-Hb), 2.88 (s, 3H, CH3),
3.53-3.70 (m, 2H, 5′-H), 3.89-3.94 (m, 1H, 4′-H), 4.40-4.50
(m, 1H, 3′-H), 5.02 (t, 1H, J ) 5.4 Hz, 3′-OH), 5.37 (d, 1H, J )
3.2 Hz, 5′-OH), 6.50 (t, 1H, J ) 6.7 Hz, 1′-H), 7.61 (d, 1H, J )
1.5 Hz, etheno Ha), 8.01 (d, 1H, J ) 1.5 Hz, etheno Hb), 8.49
(s, 1H, 8-H); m/z 312 (M + Na)+.
To a stirred solution of (NH4)2S2O8 (433 mg, 1.90 mmol) in
0.5 M phosphate buffer, pH 7.5, at room temperature was
added 6 (250 mg, 0.864 mmol). The mixture was heated at 80
°C for 1 h and then evaporated. The residue was triturated in
EtOH (100 mL) and then filtered and washed with EtOH. The
combined filtrates were evaporated to dryness, and the residue
was chromatographed on silica AcOH/MeOH/CH2Cl2, 5:20:75)
to give a white solid, which was recrystallized from EtOH and
1
Et2O to give 7 (138 mg, 60%) as a white solid. H NMR (200
MHz, DMSO-d6): δ 2.18-2.28 (m, 1H, 2′-Ha), 2.40 (s, 3H,
CH3), 2.61-2.79 (m, 1H, 2′-Hb), 3.50-3.70 (m, 2H, 5′-H), 3.90-
3.93 (m, 1H, 4′-H), 4.39-4.52 (m, 1H, 3′-H), 5.34 (t, 1H, J )
5.4 Hz, 3′-OH), 5.59 (d, 1H, J ) 3.2 Hz, 5′-OH), 6.33 (t, 1H, J
) 7.09 Hz, 1′-H), 7.26 (s, 2H, NH2), 8.26 (s, 1H, 8-H); m/z 266
(M + H)+. Anal. (C11H15N5O3‚1H2O) C, H, N.
2,N6-Dim eth yl-2′-d eoxya d en osin e (9). A mixture of dried
7 (250 mg, 0.94 mmol) and 1,2-bis[(dimethylamino)methylene]-
hydrazine (535 mg, 3.76 mmol) in anhydrous pyridine (3 mL)
was evaporated to dryness and then dried under vacuum for

970 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4
Raboisson et al.
the residue was diluted with ethyl acetate, washed with cold
water (20 mL) and saturated sodium bicarbonate (10 mL),
dried (Na2SO4), and concentrated to dryness under reduced
pressure. Chromatography on silica (CH2Cl2/MeOH, 80:20)
spectrofluorimeter (Photon Technology International Inc.,
Princetown, NJ ).4 The excitation wavelength was alternately
fixed at 340 or 380 nm, fluorescence emission was determined
at 510 nm, and results were calculated as the fluorescence
ratio (340/380) in arbitrary units.
1
afforded compound 13 (668 mg, 78%) as a colorless solid. H
NMR (300 MHz, CDCl3): δ 2.49-2.57 (m, 1H, 2′-Ha), 2.79-
2.88 (m, 1H, 2′-Hb), 3.20 (d, 3H, J ) 4.1, NCH3), 3.37-3.45
(m, 2H, 5′-H), 3.78 (s, 6H, 2OCH3), 4.11-4.15 (m, 1H, 4′-H),
4.64-4.73 (m, 1H, 3′-H), 5.73 (q, 1H, J ) 4.1, NH), 6.44 (t,
1H, J ) 6.4, 1′-H), 6.78-6.85 (m, 4H, 4CH), 7.21-7.45 (m, 9H,
9CH), 7.89 (s, 1H, 2-H), 8.35 (s, 1H, 8-H); m/z 590 (M + Na)+.
Prepared from 13 (65 mg, 0.115 mmol), potassium tert-
butoxide (1.0 M in THF, 126 µL), and tetrabenzyl pyrophos-
phate (68 mg, 0.126 mmol) using the procedure described for
compound 10, compound 14 (89 mg, 93%) was obtained as a
Mea su r em en t of Ad en ylyl Cycla se Activity. A 450 µL
aliquot of washed platelets resuspended in Tyrode’s buffer
containing 2 mM Ca2+ and 1 mM Mg2+ was stirred at 1100
rpm in an aggregometer cuvette, and the following reagents
were added at 30 s intervals: (i) 10 µM PGE1, (ii) 100 µM
compound 1e, and (iii) 5 µM ADP or vehicle (Tyrode’s buffer
containing no Ca2+ or Mg2+). The reaction was stopped 1 min
later by the addition of 50 µL of ice-cold 6.6 N perchloric acid.
Perchloric acid extracts were centrifuged at 11 000g for 5 min
to eliminate protein precipitate, and cAMP was isolated from
the supernatants using a mixture of trioctylamine and Freon
(28/22, vol/vol). The upper aqueous phase was lyophilized, and
the dry residue was dissolved in the buffer provided with the
commercial radioimmunoassay kit for cAMP measurement.4
hygroscopic colorless solid. 1H NMR (200 MHz, CDCl3):
δ
2.57-2.67 (m, 1H, 2′-Ha), 2.87-3.01 (m, 1H, 2′-Hb), 3.23 (d,
3H, J ) 4.9, NCH3), 3.27-3.70 (m, 2H, 5′-H), 3.80 (s, 6H,
2OCH3), 4.29-4.34 (m, 1H, 4′-H), 5.04-5.10 (m, 4H, 2CH2),
5.14-5.29 (m, 1H, 3′-H), 5.72 (q, 1H, J ) 4.9, NH), 6.36 (m,
1H, 1′-H), 6.76-6.83 (m, 4H, 4CH), 7.23-7.42 (m, 19H, 19CH),
7.84 (s, 1H, 2-H), 8.35 (s, 1H, 8-H); m/z 828 (M + H)+.
A stirred solution of 14 (600 mg, 0.73 mmol) in CH2Cl2 (5
mL) was treated with formic acid (5 mL) at room temperature
for 1 h. After the yellow mixture was concentrated, the residue
was redissolved in ethyl acetate (20 mL), washed with dilute
sodium bicarbonate, and dried over Na2SO4. Removal of the
solvent in vacuo afforded a residue, which was purified by
silica gel column chromatography (AcOEt/CH2Cl2/EtOH, 40:
50:10) to give compound 15 (372 mg, 97%) as a colorless solid.
1H NMR (300 MHz, CDCl3): δ 2.33-2.39 (m, 1H, 2′-Ha), 2.97-
3.10 (m, 1H, 2′-Hb), 3.18 (br s, 3H, NCH3), 3.58-3.87 (m, 2H,
5′-H), 4.23-4.25 (m, 1H, 4′-H), 5.04-5.19 (m, 5H, 3′-H, 2CH2),
6.04-6.08 (m, 1H, 1′-H), 7.18 (br s, 1H, NH), 7.30-7.45 (m,
10H, 10CH), 7.76 (s, 1H, 2-H), 8.36 (s, 1H, 8-H); m/z 526 (M +
H)+.
P oten tiom etr ic Stu d ies a n d NMR Deter m in a tion s.
Potentiometric and NMR determinations were carried out as
previously reported.35,40 The experiments were performed in
two steps in which 0.50 cm3 of the same initial solution of
compounds 1e or 1a of 5.0 × 10-3 mol dm-3 in 2H2O was
successively subjected to potentiometric and NMR titrations.
It should be noted that the glass electrode was calibrated in a
concentration scale and the measurements done in 2H2O, so
that here pH means the cologarithm of the concentration of
2H+. The processing of the pH measurements allowed the total
concentration of the ligand and the acid as well as the
macroscopic protonation constants to be determined. One-
dimensional 31P NMR spectra were recorded at 121.50 MHz
on a Bruker DPX-300 Fourier transform spectrometer. 31P
chemical shift values were referenced to an external 85%
H3PO4 signal at 0.00 ppm with downfield shifts represented
by positive values. Spectra were acquired over a spectral width
of 10 ppm using a 0.1 s relaxation delay and a π/2 pulse.
Typically, 1K data points were sampled with a corresponding
0.4 s acquisition time. Data were zero-filled and a 1 Hz
exponential line broadening function was applied prior to
Fourier transformation. The spectra had a digital resolution
of 1.19 Hz per point. The HypNMR program59 was used to
check the potentiometrically determined protonation con-
stants. The 1H NMR titration was performed on the same
equipment as before operating at 300.13 MHz. Spectra were
acquired with water presaturation over a spectral width of 6
ppm using a 3 s relaxation delay and a π/2 pulse. With a
corresponding 1.14 s acquisition time, 4K data points were
sampled . The spectra had a digital resolution of 0.44 Hz per
point. The temperature was controlled at 310 ( 0.5 K. The
proton and phosphorus resonances were assigned by perform-
ing proton-proton and phosphorus-proton 2D correlation
experiments at pH, thus allowing the titration curves to be
characterized.
A mixture of 15 (130 mg, 0.247 mmol) and 10% Pd/C (100
mg) in absolute methanol (20 mL) was shaken in a hydrogena-
tion apparatus at room temperature for 24 h. The catalyst was
removed by filtration and washed with water, and the filtrate
was concentrated to dryness. Recrystallization from methanol
and diethyl ether yielded 16 (71 mg, 83%) as a colorless solid;
1
mp 167 °C. H NMR (300 MHz, DMSO-d6): δ 2.51-2.57 (m,
1H, 2′-Ha), 2.83-2.93 (m, 1H, 2′-Hb), 2.94 (br s, 3H, CH3),
3.53-3.68 (m, 2H, 5′-H), 4.12-4.16 (m, 1H, 4′-H), 4.90-4.95
(m, 1H, 3′-H), 6.33-6.38 (m, 1H, 1′-H), 7.87 (br s, 1H, NH),
8.28 (s, 1H, 2-H), 8.34 (s, 1H, 2-H). Anal. (C11H16N5O6P‚0.5H2O)
C, H, N.
Biologica l Tests. ADP and PGE1 were from Sigma (Saint
Quentin-Fallavier, France). Human fibrinogen was from Kabi
(Stockholm, Sweden), fura-2/acetoxymethyl ester (fura-2/AM)
was from Calbiochem (Meudon, France), and the cAMP assay
kit was from Amersham (Les Ulis, France). Apyrase was
purified from potatoes as previously described.58 Compound
1a was synthesized by P. Raboisson (CNRS, Faculty of
Pharmacy, Strasbourg, France).13
Refer en ces
Wa sh ed Hu m a n P la telet Aggr ega tion . Washed human
platelets were prepared as previously described58 and resus-
pended at 3 × 105 platelets/µL in Tyrode’s buffer containing 2
mM CaCl2, in the presence of 0.02 U/mL of the ADP scavenger
apyrase (adenosine 5′-triphosphate diphosphorylase, EC 3.6.1.5).
Platelets were kept at 37 °C throughout all experiments, and
aggregation was measured by standard methods.58
(1) Ralevic, V.; Burnstock, G. Receptors for purines and pyrimidines.
Pharmacol. Rev. 1998, 50, 413-492.
(2) Hollopeter, G.; J antzen, H.-M.; Vicent, D.; Li, G.; England, L.;
Ramakrishnan, V.; Yang, R.-B.; Nurden, P.; Nurden, A.; J ulius,
D.; Conley, P. B. Identification of the platelet ADP receptor
targeted by antithrombotic drugs. Nature 2001, 409, 202-207.
(3) Gachet, C. ADP Receptors of platelets and their inhibition.
Thromb. Haemostasis 2001, 86, 222-232.
Briefly, a 450 µL aliquot of platelet suspension was stirred
at 1100 rpm and activated by the addition of agonists and
human fibrinogen (0.8 mg/mL), in a final volume of 500 µL.
The extent of aggregation was estimated quantitatively by
measuring the maximum curve height above the baseline.
[Ca 2+]i Mea su r em en ts. Fura-2/AM-loaded human platelets
were prepared as previously described4 and resuspended in
Tyrode’s buffer with 2 mM CaCl2. Aliquots of fura-2-loaded
platelets were transferred to a 10 mm × 10 mm quartz cuvette
maintained at 37 °C and fluorescence measurements were
performed under continuous stirring, in a PTI Deltascan
(4) Hechler, B.; Leon, C.; Vial, C.; Vigne, P.; Frelin, C.; Cazenave,
J . P.; Gachet, C. The P2Y1 receptor is necessary for adenosine
5′-diphosphate-induced platelet aggregation. Blood 1998, 92,
152-159.
(5) Le´on, C.; Hechler, B.; Freund, M.; Eckly, A.; Vial, C.; Ohlmann,
P.; Dierich, A.; LeMeur, M.; Cazenave, J .-P.; Gachet, C. Defective
platelet aggregation and increased resistance to thrombosis in
purinergic P2Y1 receptor-null mice. J . Clin. Invest. 1994, 104,
1731-1737.
(6) Fabre, J .-E.; Nguyen, M.; Latour, A.; Keifer, J . A.; Audoly, L.
P.; Coffman, T. M.; Koller B. H. Decreased platelet aggregation,
increased bleeding time and resistance to thromboembolism in
P2Y1-deficient mice. Nat. Med. 1999, 5, 1199-1202.

Novel Antagonists Acting at the Purinergic Receptor
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4 971
(7) Le´on, C.; Freund, M.; Ravanat, C.; Baurand, A.; Cazenave, J .-
P.; Gachet, C. Key role of the P2Y1 receptor in tissue factor-
induced thrombin-dependent acute thromboembolism studies in
P2Y1-Knockout mice and mice treated with a P2Y1 antagonist.
Circulation 2001, 103, 718-723.
(8) Baurand, A.; Raboisson, P.; Freund, M.; Le´on, C.; Cazenave, J .
P.; Bourguignon, J . J .; Gachet, C. Inhibition of platelet function
by administration of MRS2179, a P2Y1 receptor antagonist. Eur.
J . Pharmacol. 2001, 412, 213-221.
(9) Boyer, J . L.; Mohanram, A.; Gamaioni, E.; J acobson, K. A.;
Harden, T. K. Competitive and selective antagonism of P2Y1
receptors by N6-methyl-2′-deoxyadenosine-3′,5′-bisphosphate. Br.
J . Pharmacol. 1998, 124, 1-3.
(10) Boyer, J . L.; Lazarowski, E.; Chen, X.-H.; Harden, T. K.
Identification of a P2Y-purinergic receptor that inhibits adenylyl
cyclase but does not activate phospholipase C. J . Pharmacol.
Exp. Ther. 1993, 267, 1140-1146.
(11) Boyer, J . L.; Mohanram, A.; Deleney, S.; Waldo, G.; Harden, T.
K. Signaling mechanism and pharmacological selectivity of an
avian P2 receptor. Nucleotides and Their Receptors in the
Nervous System, Leipzig, Germany, Aug 1, 1998; Abstract A03.
(12) Brown, S.; King, B. F.; Kim, Y.-C.; Burnstock, G.; J acobson, K.
A. Activity of novel adenine nucleotide derivatives as agonists
and antagonists at recombinant rat P2X receptors. Drug Dev.
Res. 2000, 49, 253-259.
(13) Camaioni, E.; Boyer, J . L.; Mohanram, A.; Harden, T. K.;
J acobson, K. A. Deoxyadenosine bisphosphate derivatives as
potent antagonists at P2Y1 receptors. J . Med. Chem. 1998, 41,
183-190.
(14) Fischer, B.; Boyer, J . L.; Hoyle, C. H. V.; Ziganshin, A. U.;
Brizzolara, A. L.; Knight, G. E.; Zimmet, J .; Burnstock, G.;
Harden, T. K.; J acobson, K. A. Identification of potent, selective
P2Y-purinoceptor agonists-structure-activity-relationships for
2-thioether derivatives of adenosine 5′-triphosphate. J . Med.
Chem. 1993, 36, 3937-3946.
(15) Kim, Y.-C.; Gallo-Rodriguez, C.; J ang, S.-Y.; Nandanan, E.;
Adams, M.; Harden, T. K.; Boyer, J . L.; J acobson, K. A. Acyclic
analogues of deoxyadenosine-3′,5′-bisphosphates as P2Y1 recep-
tor antagonists. J . Med. Chem. 2000, 43, 746-755.
(30) Felemez, M.; Marwood, R. D.; Potter, B. V. L.; Spiess, B.
Inframolecular studies of the protonation of adenophostin A:
Comparison with 1-D-myo-inositol 1,4,5-trisphosphate. Biochem.
Biophys. Res. Commun. 1999, 266, 334-340.
(31) Gue´dat, P.; Schlewer, G.; Kremp, E.; Riley, A. M.; Potter, B. V.
L.; Spiess, B. Investigation of the inframolecular acid-base
properties of D-myo-inositol 1,3,4,5-tetrakisphosphate and DL-
myo-inositol 1,2,4,5-tetrakisphosphate. J . Chem. Soc., Chem.
Commun. 1997, 625-626.
(32) Liu, C. S.; Davis, R. J .; Nahorski, S. R.; Ballereau, S.; Spiess,
B.; Potter, B. V. L. Synthesis, calcium mobilizing, and physico-
chemical properties of D-chiro-inositol 1,3,4,6-tetrakisphosphate,
a novel and potent ligand at the D-myo-inositol 1,4,5-trisphos-
phate receptor. J . Med. Chem. 1999, 42, 1991-1998.
(33) Mernissi-Arifi, K.; Ballereau, S.; Schlewer, G.; Spiess, B.; Zen-
kouar, M. Microscopic acid-base properties of D-myo-inositol
1,2,6-tris(phosphate) and its 3,4,5-trideoxy analoguesinfluence
of the hydroxy groups of the myo-inositol ring. New J . Chem.
1996, 20, 1087-1092.
(34) Mernissi-Arifi, K.; Schlewer, G.; Spiess, B. Inframolecular acid-
base properties of myo-inositol 1,2,6-trisphosphate analoguess
influence of the hydroxyl groups, phosphate configuration and
intracyclic atom substitution. Carbohydr. Res. 1998, 308, 9-17.
(35) Mernissi-Arifi, K.; Schmitt, L.; Schlewer, G.; Spiess, B. Complete
resolution of the microscopic: Protonation equilibria of D-myo-
inositol 1,2,6-tris(phosphate) and related compounds by 31P NMR
and potentiometry. Anal. Chem. 1995, 67, 2567-2574.
(36) Rabenstein, D. L.; Sayer, T. L.; Determination of microscopic
acid dissociation constants by nuclear magnetic resonance
spectrometry. Anal. Chem. 1976, 48, 1141-1146.
(37) Riley, A. M.; Guedat, P.; Schlewer, G.; Spiess, B.; Potter, B. V.
L. A conformationally restricted cyclic phosphate analogue of
inositol trisphosphatessynthesis and physicochemical proper-
ties. J . Org. Chem. 1998, 63, 295-305.
(38) Sayer, T. L.; Rabenstein, D. L. Nuclear magnetic resonance
studies of the acid-base chemistry of amino acids and peptides.
III. Determination of the microscopic and macroscopic acid
dissociation constants of R,ω-diaminocarboxylic acids. Can. J .
Chem. 1976, 54, 3392-3400.
(39) Schlewer, G.; Guedat, P.; Ballereau, S.; Schmitt, L.; Spiess, B.
Inositol phosphates: inframolecular physicochemical studies,
correlation with binding properties. In Phosphoinositides: Chem-
istry, Biochemistry, and Biomedical Applications; Bruzik, K. S.,
Ed.; American Chemical Society: Washington, DC; pp 255-
270.
(40) Schmitt, L.; Bortmann, P.; Schlewer, G.; Spiess, B. Myo-inositol
1,4,5-trisphosphate and related compounds’ protonation se-
quence: potentiometric and 31P NMR studies. J . Chem. Soc.,
Perkin Trans. 2 1993, 2257-2263.
(16) Nandanan, E.; J ang, S.-Y.; Moro, S.; Kim, H. O.; Siddiqui, M.
A.; Russ, P.; Marquez, V. E.; Busson, R.; Herdewijn, P.; Harden,
T. K.; Boyer, J . L.; J acobson, K. A. Synthesis, biological activity,
and molecular modeling of ribose-modified deoxyadenosine bis-
phosphate analogues as P2Y1 receptor ligands. J . Med. Chem.
2000, 43, 829-842.
(17) Barrio, J . R.; Secrist, J . A., III; Leonard, N. J . Fluorescent
adenosine and cytidine derrivatives. Biochem. Biophys. Res.
Commun. 1972, 46, 597-604.
(18) Tsou, K. C.; Yip, K. F.; Miller, E. E.; Lo, K. W. Synthesis of 1,N6-
etheno-2-aza-adenosine (2-aza-∈-adenosine): a new cytotoxic
fluorescent nucleoside. Nucleic Acids Res. 1974, 1, 531-547.
(19) Sako, M.; Hayashi, T.; Hirota, K.; Maki, Y. A newly devised
method for the oxidative unmasking of 1,N6-ethenoadenosine:
facile conversion of adenosine into 2-deuterated adenosine.
Chem. Pharm. Bull. 1992, 40, 1656-1658.
(41) Schweizer, M. P.; Broom, A. D.; Ts’o, P. O. P.; Hollis, D. P.
Studies of inter- and intramolecular interaction in mononucle-
otides by proton magnetic resonance. J . Am. Chem. Soc. 1968,
90, 1042-1055.
(42) Martin, R. B. Nucleoside sites for transition metal ion binding.
Acc. Chem. Res. 1985, 18, 32-38.
(43) Blindauer, C. A.; Holy, A.; Dvorakova, H.; Sigel, H. Solution
properties of antiviral adenine-nucleotide analoguessthe acid-
base properties of 9-[2-(phosphonomethoxy)ethyl]adenine (pmea)
and of its n1, n3 and n7 deaza derivatives in aqueous solution.
J . Chem. Soc., Perkin Trans. 2 1997, 2353-2363.
(20) Barlett, R. K.; Humphrey, I. R. Transamination of N,N-dimeth-
ylformamideazine. J . Chem. Soc. 1967, 17, 1664-1666.
(21) Samano, V.; Miles, R. W.; Robins, M. J . Efficient conversion of
6-aminopurines and nucleosides into 6-substituted analogues via
novel 6-(1,2,4-triazol-4-yl)purine derivatives. J . Am. Chem. Soc.
1994, 116, 9331-9332.
(22) Vacca, J . P.; deSolms, S. J .; Huff, J . R. Total synthesis of D- and
L-myo-inositol 1,4,5-trisphosphate. J . Am. Chem. Soc. 1987, 109,
3478-3479.
(23) Yu, K.-L.; Fraser-Reid, B. A novel reagent for the synthesis of
myo-inositol phosphates: N,N-diisopropyl dibenzyl phosphora-
midine. Tetrahedron Lett. 1988, 29, 979-982.
(24) Ti, G. S.; Gaffney, B. L.; J ones, R. A. Transient protection:
efficient one-flask syntheses of protected deoxynucleosides. J .
Am. Chem. Soc. 1982, 104, 1316-1319.
(25) Taunton-Rigby, A.; Kim, Y.-H.; Crosscup, C. J .; Starkovsky, N.
A. Oligonucleoside synthesis. II. The Use of substituted trityl
groups. J . Org. Chem. 1972, 37, 956-964.
(26) Humphries, R. G. Pharmacology of AR-C69931MX and related
compounds: from pharmacological tools to clinical trials. Hae-
matologica 2000, 85, 66-72.
(27) Martell, A. E.; Motekaitis, R. J . Determination and Use of
Stability Constants, 2nd ed.; VCH Publishers: New York, 1992.
(28) Sarneski, J . E.; Reilley, C. N. The determination of proton-
binding sites by “nmr titrations”. In Essays on Analytical
Chemistry; Wa¨nninen, E., Ed.; Pergamon Press: Oxford, 1977;
pp 35-49.
(29) Felemez, M.; Bernard, P.; Schlewer, G.; Spiess, B. Inframolecu-
lar protonation process of myo-inositol 1,4,5-tris(phosphate) and
related compounds: Dynamics of the intramolecular interactions
and evidence of C-H‚‚‚O hydrogen bending. J . Am. Chem. Soc.
2000, 122, 3156-3165.
(44) Tribolet, R.; Sigel, H. Self-association and protonation of ad-
enosine 5′-monophosphate in comparison with its 2′- and 3′-
analogues and tubericin 5′-monophosphate (7-deaza-AMP). Eur.
J . Biochem. 1987, 163, 353-363.
(45) Wang, X.; Simpson, J . H.; Nelson, D. J . 1H and 31P NMR study
of speciation in systems containing ADP, Al3+, and fluoride. J .
Inorg. Biochem. 1995, 58, 29-47.
(46) White, A. M.; Varney, M. A.; Watson, S. P.; Rigby, S.; Chang-
sheng, L.; Ward, J . G.; Reese, C. B.; Graham, H. C.; Williams,
R. J . P. Influence of Mg2+ and pH on nmr spectra and radioligand
binding of inositol 1,4,5-trisphosphate. Biochem. J . 1991, 278,
759-764.
(47) Halbfinger, E.; Major, D. T.; Ritzmann, M.; Ubl, J .; Reiser, G.;
Boyer, J . L.; Harden, K. T.; Fischer, B. Molecular recognition of
modified adenine nucleotides by the P2Y(1)-receptor. 1.
A
synthetic, biochemical, and NMR approach. J . Med. Chem. 1999,
42, 5325-5337.
(48) Thibaudeau, C.; Plavec, J .; Chattopadhyaya, J . Quantitation of
the anomeric effect in adenosine and guanosine by comparison
of the thermodynamics of the pseudorotational equilibrium of
the pentofuranose moiety in N- and C-nucleosides. J . Am. Chem.
Soc. 1994, 116, 8033-8037.
(49) Plavec, J .; Garg, N.; Chattopadhyaya, J . How does the steric
effect drive the sugar conformation in the 3′-C-branched nucleo-
sides? J . Chem. Soc., Chem. Commun. 1993, 1011-1014.

972 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 4
Raboisson et al.
(50) Thibaudeau, C.; Plavec, J .; Watanabe, K. A.; Chattopadhyaya,
J . How do the aglycones drive the pseudorotational equilibrium
of the pentofuranose moiety in C-nucleosides? J . Chem. Soc.,
Chem. Commun. 1994, 537-540.
(51) Plavec, J .; Thibaudeau, C.; Viwanadham, G.; Sund, C.; Chatto-
padhyaya, J . How does the 3′-phosphate drive the sugar
conformation in DNA? J . Chem. Soc., Chem. Commun. 1994,
781-783.
(52) Haasnoot, C. A. G.; De Leeuw, F. A. A. M.; Altona, C. The
relation between proton-proton NMR coupling constants and
substituent electronegativities. I. An empirical generalization
of the Karplus equation. Tetrahedron 1980, 36, 2783-2792.
(53) Beevers, A. P. G.; Witch, E. M.; J ones, B. C. N. M.; Cosstick, R.;
Arnold, J . R. P.; Fisher, J . Conformational analysis of 3′-S-PO3-
linked ribo- and deoxyribodinucleoside monophosphates. Magn.
Reson. Chem. 1999, 37, 814-820.
of adenophostin, a very potent Ca2+ inducer, at the D-myoinositol
1,4,5-trisphosphate receptor. Biochemistry 1999, 38, 9234-9241.
(56) Plavec, J .; Thibaudeau, C.; Chattopadhyaya, J . How does the
2′-hydroxy group drive the pseudorotational equilibrium in
nucleoside by the tuning of the 3′-gauche effect? J . Am. Chem.
Soc. 1994, 116, 6558-6560.
(57) Hofmann, M.; Adeco, M.; Fagan, P.; Wemmer, D.; Eritja, R.; Diaz,
A. R. Synthesis of oligodeoxynucleotides containing N6-[13C-
methyl]adenine and N2-[13C-methyl]guanine. J . Chem. Soc.,
Perkin Trans. 1 1997, 12, 1825-1828.
(58) Cazenave, J . P.; Hemmendinger, S.; Beretz, A.; Sutter-Bay, A.;
Launay, J . L’agre´gation plaquettaire: outil d’investigation cli-
nique et d’e´tude pharmacologiques-methodologie. Ann. Biol.
Clin. 1983, 41, 167-179.
(59) Frassineti, C.; Ghelli, S.; Gans, P.; Sabatini, A.; Moruzzi, M. S.;
Vacca, A. Nuclear magnetic resonance as a tool for determining
protonation constants of natural polyprotic bases in solution.
Anal. Biochem. 1995, 231, 374-382.
(54) Altona, C.; Sundaralingam, M. Conformational analysis of the
sugar ring in nucleosides. Improved for the interpretation of
proton magnetic resonance coupling constants. J . Am. Chem.
Soc. 1973, 95, 2333-2344.
(55) Hotoda, H.; Murayama, K.; Miyamoto, S.; Iwata, Y.; Takahashi,
M.; Kawase, Y.; Tanzawa, K.; Kaneko, M. Molecular recognition
J M0104062