Antagonists of the platelet P(2t) receptor: A novel approach to antithrombotic therapy

Article abstract of DOI:10.1021/jm981072s

The platelet P(2T) receptor plays a major role in platelet aggregation, and its antagonists are predicted to have significant therapeutic potential as antithrombotic agents. We have explored analogues of adenosine triphosphate (ATP), which is a weak, nonselective but competitive P(2T), receptor antagonist. Modification of the polyphosphate side chain to prevent breakdown to the agonist adenosine diphosphate (ADP) and substitution of the adenine moiety to enhance affinity and selectivity for the P(2T) receptor led to the identification of 10e (AR-C67085MX), having an IC₅₀ of 2.5 nM against ADP-induced aggregation of human platelets. Compound 10e was the first very potent antagonist of the P(2T) receptor, with a selectivity for that subtype of the P2 receptor family of > 1000-fold. Further modification of the structure produced compound 101 (AR-C69931MX) having an IC₅₀ of 0.4 nM. In vivo, at maximally effective antithrombotic doses, there is little prolongation of bleeding time (1.4-fold), which is in marked contrast to the 5-6 fold found with GPIIb/IIIa antagonists.

Full text of DOI:10.1021/jm981072s

J . Med. Chem. 1999, 42, 213-220
213
Expedited Articles
An ta gon ists of th e P la telet P 2T Recep tor : A Novel Ap p r oa ch to An tith r om botic
Th er a p y
Anthony H. Ingall,* J ohn Dixon, Andrew Bailey, Mandy E. Coombs, David Cox, J udith I. McInally,
Simon F. Hunt, Nicholas D. Kindon, Barry J . Teobald, Paul A. Willis, Robert G. Humphries,† Paul Leff,†
J ane A. Clegg,† J ames A. Smith,† and Wendy Tomlinson†
Departments of Medicinal Chemistry and Pharmacology, ASTRA Charnwood, Bakewell Road, Loughborough LE11 5RH, U.K.
Received August 11, 1998
The platelet P2T receptor plays a major role in platelet aggregation, and its antagonists are
predicted to have significant therapeutic potential as antithrombotic agents. We have explored
analogues of adenosine triphosphate (ATP), which is a weak, nonselective but competitive P2T
receptor antagonist. Modification of the polyphosphate side chain to prevent breakdown to the
agonist adenosine diphosphate (ADP) and substitution of the adenine moiety to enhance affinity
and selectivity for the P2T receptor led to the identification of 10e (AR-C67085MX), having an
IC50 of 2.5 nM against ADP-induced aggregation of human platelets. Compound 10e was the
first very potent antagonist of the P2T receptor, with a selectivity for that subtype of the P2
receptor family of >1000-fold. Further modification of the structure produced compound 10l
(AR-C69931MX) having an IC50 of 0.4 nM. In vivo, at maximally effective antithrombotic doses,
there is little prolongation of bleeding time (1.4-fold), which is in marked contrast to the 5-6-
fold found with GPIIb/IIIa antagonists.
In tr od u ction
and P2t.1,2 Moreover there was a complete dearth of
information on the physiological importance of P2
receptors in man. The application of molecular biological
techniques in the last 5 years has greatly expanded the
number of variants within the P2 receptor family while
simplifying its structure, and the nomenclature has
been modified accordingly. Thus the “P2y receptor”
comprises a group of transmembrane helix G-protein-
coupled receptors (designated P2Y1 ,etc.), and the “P2x
receptor” has been subdivided into eight ion channels
(P2X1-8), subsuming the P2z type. Many of the members
of these subfamilies have been cloned and sequenced.3
As yet there is no definitive structural information
on the P2T receptor, consequently its true family rela-
tionship is unclear, though recent evidence that it is
G-protein-coupled4 suggests it belongs in the P2Y
subclass. While other P2 receptors have wide tissue
distributions, the P2T subtype has, to date, only been
found on platelets. It has recently been shown that ADP
stimulates platelets via its action on three distinct P2
receptors (P2X1, P2Y1, and P2T) leading to shape change,
exposure of the glycoprotein IIb/IIIa complex (binding
sites for cross-linking of platelets by fibrinogen), am-
plification of the response by release of a range of
proaggregatory mediators (including thromboxane A2,
5-HT, and ADP), and sustained aggregation. Studies
with the novel selective compounds described herein
indicate a pivotal role for the P2T receptor in the
aggregatory process, as all of these events, with the
exception of shape change, are blocked by P2T receptor
antagonists.
Platelet adhesion and aggregation are pivotal events
in arterial thrombosis. Activated under conditions of
turbulent blood flow in diseased vessels or by the release
of mediators from other circulating cells and damaged
endothelial cells lining the vessel, platelets accumulate
at a site of vessel injury and recruit further platelets
into the developing thrombus. As a component of normal
hemostasis, platelet aggregation serves a vital physi-
ological function by arresting bleeding following trau-
matic transection of blood vessels. However, platelet
aggregation at the site of a ruptured atherosclerotic
plaque on the surface of, for example, a coronary artery
may result in life-threatening partial or complete oc-
clusive arterial thrombosis manifesting as unstable
angina or myocardial infarction. Intervention in the
process of thrombus formation has long been an attrac-
tive therapeutic target for the treatment or prophylaxis
of such events, and many approaches, centered on the
various components of the process, have been investi-
gated.
Our interest in antithrombotic therapy stemmed from
our investigations of P2 receptors (a receptor family
whose endogenous agonists are nucleotides) and the
working hypothesis that activation of the P2T receptor
subtype on platelets by adenosine diphosphate (ADP)
is a key event in arterial thrombosis. At the time of our
entry into the area, four subtypes of the P2 family were
known and, in the absence of structural information and
selective antagonists, were then designated P2x, P2y, P2z,
The original basis for the chemical program was the
observation that ATP is a competitive, albeit weak,
* To whom correspondence should be addressed.
† Department of Pharmacology.
10.1021/jm981072s CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/28/1999

214 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2
Ingall et al.
Sch em e 1a
a
Reagents: (a) isoamyl nitrite, MeCN, electrophile (e.g., RSSR, CH2I2); (b) R6NH2.
Sch em e 2a
a
Reagents: (a) CF3CH2CH2I, NaOH, H2O; (b) Ac2O, NaOAc; (c) NaH, DMF, R6Hal; (d) MeO-, MeOH, heat.
antagonist of the ADP-induced platelet aggregation.
Coupling this information with the exclusive localization
of the P2T receptor on the platelet, we concluded that
potent and selective ATP analogue antagonists of this
receptor would offer great potential as novel antithrom-
botic agents.
Elaboration of the phosphorus-containing side chain
was by the method of Scheme 3. Phosphorylation of the
5′-hydroxyl of the modified adenosine moiety 3 (or 5)
by the method of Yoshikawa9 afforded the 5′-monophos-
phates 8 efficiently, which, after removal of the phos-
phoric acid byproduct by ion-exchange chromatography
on a sulfonic acid resin (Dowex 50W (H+ form)) (using
aqueous ammonia as eluant), could be activated by
conversion to phosphorimidazolates 910 with carbonyl-
diimidazole. This unstable intermediate (generated in
situ and not isolated) was coupled with the methyl-
enebis(phosphonic acids). Ion-exchange chromatography
on weakly basic DEAE Sephadex (using aqueous tri-
ethylammonium bicarbonate or ammonium bicarbonate
as eluant) was the preferred method of purification of
the “stabilized triphosphate products” 10. Separation of
the product and unreacted monophosphate residues was
easily monitored by UV detection of the chromophoric
components. However separation of the product from
the excess of the (non-UV-active) methylenebis(phos-
phonic acids) could not be determined by the same
means and was checked by 31P NMR after the product-
containing fractions had been concentrated by freeze-
drying. The sensitivity of the 31P nucleus in NMR
experiments was sufficiently high as to permit a good
determination of “phosphorus purity” of the chromato-
graphic fractions, which were combined as appropriate
and freeze-dried again, affording the required products
as (alkyl)ammonium salts. These salts were frequently
very hygroscopic or even deliquescent; so in the majority
of cases stable solid sodium salts were prepared by
Ch em istr y
Two general methods were used to prepare the
modified adenosine precursors of the target structures,
the choice of route being dictated by the accessibility of
the necessary reagents for the introduction of the C-2
or N6 substituent. Method A (Scheme 1) was preferred
as offering the broadest possible range of combinations
of substituents. Diazotization of the amino group of 1
(prepared from guanosine5) with isoamyl nitrite led to
the purinyl radical which could be efficiently trapped
by disulfides,6 halogen sources,7 or other electrophilic
agents, giving 2. Subsequent displacement of the chlo-
rine atom at C-6 by ammonia or amines conveniently
also removed the acetyl protecting groups to afford the
modified nucleosides 3c-h .
For the preparation of the 3,3,3-trifluoropropylthio-
substituted compounds 3i-l (see Table 1) method B
(Scheme 2) was used. 2-Mercaptoadenosine8 (4) alky-
lated readily on the thiol group. Acetylation of the
S-alkylated material 5 leading to 6 simultaneously
protected the ribose hydroxyls and activated the exo-
cyclic amino group to deprotonation with sodium hy-
dride. This anion could be trapped with a range of
electrophiles, affording 7; subsequent mild treatment
with ammonia or methoxide removed the protecting
groups to give the adenosine analogues 3i-l.

Antagonists of the Platelet P2T Receptor
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2 215
Sch em e 3a
a
Reagents: (a) POCl3, (EtO)3PO; (b) CDI; (c) R2C(PO3H2)2.
Ta ble 1. P2T Receptor Antagonist Activity (Human Washed Platelets) and Recovery of Aggregation Behavior 20 min Postadministra-
tion in Rat
20 min
recovery mean
a
compd
R
R2
R6
pIC50
3.5
% (spread)b
10a c
10bc
10c
10d
10e
10f
10g
10h
10i
F
H
H
H
H
H
H
H
Cl
Cl
F
3.5
SEt
SPr
SPr
SPr
SPr
SPr
6.53 ( 0.16
8.16 ( 0.19
8.60 ( 0.09
9.48 ( 0.06
7.81 ( 0.18
9.42 ( 0.11
8.87 ( 0.09
9.14 ( 0.11
8.02 ( 0.15
9.35 ( 0.21
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
93 (81-100)
10 (3-16)
48 (43-56)
33 (21-43)
CH2CF3
CH2CH2OMe
CH2CH2SMe
H
CH2CF3
CH2CH2OMe
CH2CH2SMe
SCH2CH2CF3
SCH2CH2CF3
SCH2CH2CF3
SCH2CH2CF3
10j
10k
10l
6 (1-11)
64 (54-79)
79 (60-107)
a
b
Mean of four determinations except for 10l (nine determinations). Recovery of ADP-induced platelet aggregation 20 min postdose;
mean of four determinations and spread of results. c Reference 15.
metathesis with sodium iodide. Even then the salts were
heavily hydrated, and they were used as amorphous
powders.
tors.) Table 1 reports the pIC50 data for a selection of
key compounds.
ATP, as a starting point for the discovery of a useful
antithrombotic agent, presents a range of intriguing
challenges: chemical and biological lability of the poly-
phosphate moiety, low affinity for the P2T receptor (pIC50
3.6), lack of selectivity between P2 subtypes, and
extreme physicochemical properties (log D7.4 , -3).
There is an unmet therapeutic need for an intravenous
antithrombotic agent for use in episodes of acute arterial
thrombosis. In such circumstances the highly polar
nature of analogues of ATP could be advantageous, and
we have identified such a compound that could be used
for the treatment of acute thrombotic conditions.
Enzymatic or chemical hydrolysis of ATP is facile,
leading to ADP (an endogenous proaggregatory agent),
Resu lts a n d Discu ssion
Human washed platelets were used for the primary
screen in vitro. Suspended in a modified Tyrode buffer,
they were stimulated to aggregate by the addition of
ADP, and the aggregation process was detected using
a turbidimetric technique.11 Aggregation was inhibited
by the test compounds in a concentration-dependent
manner, from which a pIC50 value could be calculated.
(Duplicate experiments were conducted in the presence
of the A2a adenosine receptor antagonist 8-(sulfophenyl)-
theophylline to identify effects mediated via P1 recep-

216 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2
Ingall et al.
AMP, and adenosine, which, inter alia, can also inhibit
aggregation, via the A2A receptor on platelets. Enhance-
ment of the stability of the polyphosphate side chain to
prevent the formation of the proaggregatory diphos-
phate was the most important preliminary target. The
ectonucleotidases that are found on the surface of many
cardiovascular tissues generally degrade polyphosphate
species by stepwise removal of terminal phosphates.
Replacement of the anhydride oxygen between Pâ and
Pγ (the central and terminal phosphorus atoms) of ATP
by a methylene unit has been shown to afford a
degradation-resistant product,12 though with an impor-
tant effect upon the ionization behavior of the triphos-
phate. The pKa for the final deprotonation of ATP is
∼6.6, while for the âγ-methylene compound it is ∼8.1.
At physiological pH, therefore, ATP may be expected
to be predominantly tetra-anionic and the âγ-methylene
analogue predominantly a tri-anion. Enhancement of
the electronegativity of the âγ-carbon unit by halogena-
tion gives rise to compounds with pKa values close to
those of ATP;13 this maneuver has been used extensively
by Blackburn et al. in investigating the role nucleotides
play in various biological processes14 and in the early
general investigation of the P2 receptors by Cusack and
Hourani.15
the P2T receptor had little effect on the potencies at the
other P2 receptors (ATP is an agonist in these tissues),
and a selectivity of some >104-fold was found for 10e.
Being the first example of a highly potent and
selective antagonist of any P2 purinoceptor, 10d (AR-
C66096MX, formerly FPL 66096MX) was recognized as
an early definitive pharmacological tool for character-
izing P2T receptors.11 The later compound AR-C67085MX
(10e), having slightly superior receptor affinity and
containing the preferred dichloromethylenebis(phospho-
nic acid) moiety, was taken forward into development
for the express purpose of permitting studies of the
significance of a P2 receptor in humans.
In vivo studies showed that 10e had a very short
duration of action, in the anesthetized rat, dog, and
human (t1/2 in human approximately 2 min).20 In a dog
model of arterial thrombosis, a full antithrombotic effect
was achieved at a dose (0.1 nmol‚kg-1‚min-1 iv) which
produced complete inhibition of platelet aggregation,
measured ex vivo, and no detectable hemodynamic
effects.21 Full inhibition of aggregation was established
within 15 min of commencement of infusion, and normal
baseline aggregation was fully restored within 15 min
of cessation of dosing. These onset/offset parameters
were maintained even following continuous infusion of
the compound for 6 h or more. This highly selective,
rapid onset/rapid offset profile was felt to be highly
desirable in the acute care setting.
While dialkylation of N6 of the adenine reduced
receptor affinity (the N6,N6-diethyl derivative of 10e was
some 1000-fold less potent than 10e), monoalkylation
gave a useful increase. A lipophilic substituent was
essential (polar atoms in or upon the chain reduced
receptor binding markedly: see, for example, compound
10g in Table 1), and activity increased with chain
length. However, there was an effective upper limit to
the length that could be tolerated, as chains of ap-
proximately four carbon atoms and longer caused the
compound’s biological half-life to increase sharply, unac-
ceptably compromising the rapid onset/offset profile
shown by 10e in the dog. Differences in the functional
kinetics of compounds were subsequently explored in
an anesthetized rat model,22 in which ADP-induced ex
vivo aggregation of blood samples taken at various time
points after iv administration of compound was used to
follow recovery of aggregatory behavior and hence
clearance of compound. In this screen 10e showed >93%
recovery 20 min after dosing. Long substituent chains
on N6 reduced this to <10%, as, surprisingly, did a 2,2,2-
trifluoroethyl group (see 10f,j). A methylthioethyl sub-
stituent at this position (10h ) was found to enhance
affinity with somewhat less effect upon half-life. In
combination with a 3,3,3-trifluoropropylthio group at
C-2, the N6-methylthioethyl group gave a compound
with a half-life similar to that of 10e (showing 79%
recovery of aggregation behavior at the 20 min time
point). This compound 10l, being some 6-fold more
potent than 10e, has been taken forward into develop-
ment and is currently in phase II of development.
We hypothesized that the agonist/antagonist behavior
difference of ADP and ATP was the consequence of
charge density in the region of space occupied by the
terminal phosphate of ATP, which is absent in ADP.
The stabilized ATP analogues were all found to be
antagonists, but the âγ-dihalomethylene compounds
10a ,b (having pKa values similar to that of ATP) had
higher affinities (pIC50 3.5) than the methylene com-
pound. While the dichloro- and difluoro-substituted
structures had comparable activities, the dichloro-
substituted system was preferred, since dichlorometh-
ylenebis(phosphonic acid) (clodronic acid) has been
extensively investigated as a pharmaceutical per se and
appears to be toxicologically acceptable;16 moreover its
synthesis is simpler and safer than its fluorinated
analogue.
Substitutions of C-8 and C-2 of the adenine group
were next investigated. The introduction of even small
groups at C-8 produced a reduction of affinity (data not
shown), while replacement of the hydrogen at C-2 led
to important gains in antagonist activity. Taken to-
gether these observations suggest the required confor-
mation about the glycosidic bond to be “anti”. Although
anything other than hydrogen at C-2 conferred en-
hanced activity, nonpolar moieties were preferred and
sulfide-linked chains were particularly advantageous,
an effect previously observed by Gough et al.17 who
noted that 2-methylthio-ADP was some 30-fold more
potent as a platelet-aggregating agent than ADP. In-
vestigation of the homologous series of alkyl sulfide
chains revealed a quite unexpected 100-fold leap of
affinity between the S-ethyl and S-propyl examples (see
compounds 10c,e, Table 1); neither further increase in
chain length nor substitution found any additional
substantial advantage. Selectivity was initially assessed
using functional P2X and P2Y assays in rabbit ear
artery18 and guinea pig aorta19 preparations, respec-
tively, and subsequently in binding assays using cloned
receptors. The modifications that enhanced affinity for
While profound inhibition of platelet function provides
the potential for beneficial antithrombotic effects, this
may be achieved at the expense of compromised hemo-
stasis, with a subsequent increase in the risk of bleed-
ing, particularly during invasive procedures. To assess

Antagonists of the Platelet P2T Receptor
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2 217
F igu r e 1. Effects of compound 10l (AR-C69931MX) (panel a, 5-6 dogs) and the GPIIb/IIIa antagonist Lamifiban (Ro 449883)
(panel b, 5 dogs) on femoral artery cyclical flow reduction (b), ADP-induced platelet aggregation ex vivo (0), and bleeding time
(tongue) (∆) of the anesthetized dog. Values are the mean ( SEM. Lamifiban from reference 23.
the potential for undesirable antihemostatic effects, we
have examined the relationship between the antithrom-
botic activity of 10l and its effect upon bleeding time in
the dog. Dose-response relationships show a favorable
(98-fold) separation between the desired antithrombotic
action and the prolongation of bleeding time. In conse-
quence, the full inhibition of platelet aggregation needed
to give an antithrombotic effect is achieved at doses
which extend bleeding time by less than 2-fold (see
Figure 1a). This substantial separation of the two effects
is in marked contrast to the pharmacological behavior
of members of the GPIIb/IIIa antagonist class (agents
currently attracting considerable interest for antithrom-
botic therapy). We have examined three examples (of
diverse structure and profile), and each has shown a
separation of a mere 2-5-fold. Figure 1b shows the data
generated with the compound Lamifiban (Ro 449883),
which is quite representative of our observations. Full
inhibition of aggregation by these GPIIb/IIIa antago-
nists is achieved at dose levels such that bleeding time
is prolonged by some 6-7-fold. Clearly the risk of
persistent hemorrhage following administration of a P2T
receptor antagonist is likely to be very much less than
would be the case with such GPIIb/IIIa antagonists. We
believe this substantial separation of the two effects
constitutes a major advantage for the P2T antagonist
mechanism of action in the modulation of platelet
aggregation.
On the basis of this clear and consistent advantage, we
believe antagonists of the P2T receptor represent a major
step forward in the treatment of thrombotic disease.
Exp er im en ta l Section
Thin layer chromatographic (TLC) analyses were performed
using Whatman silica gel F254 plates; visualization used UV
light or an iodine chamber. Flash chromatography for routine
purification of reaction products used silica gel (230-400
mesh). 1H and 31P NMR spectra were recorded on a Bruker
AM360 spectrometer at 360.13 and 145.73 MHz, respectively.
1H chemical shifts are expressed in ppm relative to TMS as
an internal standard, while 31P shifts are quoted relative to
85% phosphoric acid (external standard). Mass spectra were
obtained on a VG 70-250SEQ spectrometer using fast atom
bombardment ionization (FAB). Combustion analyses were
performed using a Carlo Erba EA 1108 elemental analyzer.
Water content was determined using a Orion AF7LC Karl
Fischer titrater.
Gen er a l P r oced u r es for th e P r ep a r a tion of N6-Alk yl-
2-(a lk ylth io)a d en osin es 3. 1. Meth od A, Sch em e 1. (a )
2-(Alk ylt h io)-6-ch lor o-9-(2,3,5-t r i-O-a cet yl-â-D-r ib ofu r a -
n osyl)-9H-p u r in es 2. A solution of the guanosine derivative
1 (25 g, 58.5 mmol), isoamyl nitrite (42.51 g, 363 mmol), and
the appropriate dialkyl disulfide (585 mmol) in dry acetonitrile
(350 mL) was purged thoroughly with nitrogen and then
heated at 60 °C for 20 h. The disappearance of starting
material was monitored by TLC (EtOAc-60/80 petroleum
ether, 1:1). Solvent and other volatiles were removed under
reduced pressure, and the residue was purified by column
chromatography (EtOAc-60/80 petroleum ether, 1:1 as elu-
ant).
6-Ch lor o-2-(eth ylth io)-9-(2,3,5-tr i-O-acetyl-â-D-r ibofu r a-
n osyl)-9H-p u r in e (2c). Prepared from 1 and diethyl disulfide
in 64% yield: MS (FAB) m/z 473/5 (M+ + H+), 259 (100%); 1H
NMR δ(CDCl3) 8.10 (1H, s, H-8), 6.12 (1H, d, J ) 5.3 Hz, H-1′),
5.94 (1H, t, J ) 5.5 Hz, H-2′), 5.65 (1H, t, J ) 5.5 Hz, H-3′),
4.42-4.31 (3H, m, H-4′ + H-5′a + H-5′b), 3.24 (2H, m, SCH2),
2.15 (3H, s, CH3CO), 2.12 (3H, s, CH3CO), 2.11 (3H, s, CH3CO),
1.12 (3H, t, SCH2CH3).
Con clu sion
Modification of ATP has led to the identification of
10d (AR-C66096MX), an important pharmacological tool
for the characterization of P2T receptors, and to the
discovery of the therapeutically useful analogues 10e
(AR-C67085MX) and 10l (AR-C69931MX). In a healthy
human volunteer study with 10e, a dose-related inhibi-
tion of ex vivo ADP-induced platelet aggregation has
been demonstrated, with rapid recovery on termination
of the infusion and with little or no prolongation of
bleeding time. The same profile has been demonstrated
for 10l, which is in phase II development for the
treatment of acute coronary syndromes such as unstable
angina and in the setting of percutaneous transluminal
coronary artery revascularization. A significant feature
of the pharmacological profile of P2T receptor antago-
nists is the unprecedentedly advantageous separation
of antithrombotic activity from effects on bleeding time.
6-Ch lor o-2-(p r op ylt h io)-9-(2,3,5-t r i-O-a cet yl-â-D-r ib o-
fu r a n osyl)-9H-p u r in e (2d ). Prepared from 1 and dipropyl
disulfide in 59% yield: MS (FAB) m/z 487/9 (M+ + H+), 259
1
(100%); H NMR δ(CDCl3) 8.11 (1H, s, H-8), 6.14 (1H, d, J )
5.3 Hz, H-1′), 5.92 (1H, t, J ) 5.5 Hz, H-2′), 5.60 (1H, t, J )
5.6 Hz, H-3′), 4.47-4.32 (3H, m, H-4′ + H-5′a + H-5′b), 3.20
(2H, t, J ) 7.1 Hz, SCH2), 2.15 (3H, s, CH3CO), 2.12 (3H, s,
CH3CO), 2.11 (3H, s, CH3CO), 1.81 (2H, m, SCH2CH2CH3), 1.07
(3H, t, J ) 7.3 Hz, SCH2CH2CH3).
(b) N6-Alk yl-2-(a lk ylth io)a d en osin es 3f-h . The chloro-
purine 2 (33 mmol) was suspended in 1:1 aqueous 1,4-dioxane
(400 mL) and treated with the appropriate amine (230 mmol)
at 95 °C for 24 h. The reaction was monitored by TLC (EtOAc-

218 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2
Ingall et al.
MeOH, 15:1). Solvent was evaporated under reduced pressure,
and the residue was purified by column chromatography
(EtOAc-MeOH, 15:1).
was stirred under a nitrogen atmosphere at 60 °C for 24 h.
Disappearance of starting material was monitored by TLC
(EtOAc-cyclohexane, 1:1). Solvent was removed under re-
duced pressure, and the residue was taken in EtOAc (300 mL)
and washed with water (3 × 100 mL). Drying and evaporation
gave the crude product which was purified by column chro-
matography (EtOAc-cyclohexane, 1:1).
2-(P r op ylth io)-N6-(2,2,2-tr iflu or oeth yl)a d en osin e (3f).
Prepared from 2d and 2,2,2-trifluoroethylamine in 72% yield:
MS (FAB) m/z 424 (M+ + H+), 292 (100%); 1H NMR δ(DMSO-
d6) 8.58 (1H, br s, NH), 8.34 (1H, s, H-8), 5.84 (1H, d, J ) 5.7
Hz, H-1′), 5.47 (1H, d, J ) 6.2 Hz, OH), 5.21 (1H, d, J ) 4.9
Hz, OH), 5.07 (1H, t, J ) 5.5 Hz, OH), 4.59 (1H, q, J ) 5.7 Hz,
H-2′), 4.29 (2H, br, NCH2), 4.14 (1H, q, J ) 4.9 Hz, H-3′), 3.93
(1H, m, H-4′), 3.65 (1H, m, H-5′a), 3.54 (1H, m, H-5′b), 3.06
(2H, m, SCH2), 1.71 (2H, m, SCH2CH2CH3), 0.99 (3H, t, J )
7.5 Hz, SCH2CH2CH3).
The alkylated material 7 (7.5 mmol) was dissolved in a
solution of 0.1 M sodium hydroxide in methanol (150 mL) and
heated under reflux for 30 min. After cooling the solution was
acidified with glacial AcOH (0.89 mL, 15 mmol) and evapo-
rated to dryness. The crude product was purified by column
chromatography (CH2Cl2-MeOH, 95:5).
N6-(2-Meth oxyeth yl)-2-(p r op ylth io)a d en osin e (3g). Pre-
pared from 2d and 2-methoxyethylamine in 87% yield: MS
N6-(2,2,2-Tr iflu or oeth yl)-2-(3,3,3-tr iflu or op r op ylth io)-
a d en osin e (3j). Prepared from 6 and 2,2,2-trifluoroethyl
1
1
(FAB) m/z 400 (M+ + H+, 100%); H NMR δ(DMSO-d6) 8.23
iodide in 55% yield: MS (FAB) m/z 478 (M+ + H+, 100%); H
(1H, s, H-8), 7.89 (1H, br s, NH), 5.81 (1H, d, J ) 5.7 Hz, H-1′),
5.45 (1H, d, J ) 6.3 Hz, OH), 5.20 (1H, d, J ) 4.85 Hz, OH),
5.09 (1H, t, J ) 5.4 Hz, OH), 4.60 (1H, m, H-2′), 4.12 (1H, m,
H-3′), 3.91 (1H, m, H-4′), 3.90-3.43 (6H, m, H-5′a + H-5′b +
NCH2 + OCH2), 3.28 (3H, s, OCH3), 3.07 (2H, m, SCH2), 1.70
(2H, m, SCH2CH2CH3), 0.99 (3H, t, J ) 7.3 Hz, SCH2CH2CH3).
NMR δ(DMSO-d6) 8.56 (1H, br s, NH), 8.30 (1H, s, H-8), 5.74
(1H, d, J ) 6 Hz, H-1′), 5.41 (1H, d, J ) 6.1 Hz, OH), 5.21
(1H, d, J ) 5.1 Hz, OH), 5.01 (1H, t, J ) 5.5 Hz, OH), 4.57
(1H, m, H-2′), 4.33 (2H, br, NCH2), 4.19 (1H, m, H-3′), 3.91
(1H, m, H-4′), 3.63-3.58 (2H, m, H-5′a + H-5′b), 3.20 (2H, m,
SCH2), 2.71 (2H, m, SCH2CH2CF3).
N6-(2-Meth ylth ioeth yl)-2-(p r op ylth io)a d en osin e (3h ).
Prepared from 2d and 2-methylthioethylamine in 76% yield:
MS (FAB) m/z 416 (M+ + H+, 100%); 1H NMR δ(DMSO-d6)
8.28 (1H, s, H-8), 7.81 (1H, br s, NH), 5.82 (1H, d, J ) 5.6 Hz,
H-1′), 5.47 (1H, d, J ) 6.1 Hz, OH), 5.18 (1H, d, J ) 4.8 Hz,
OH), 5.03 (1H, t, J ) 5.5 Hz, OH), 4.65 (1H, m, H-2′), 4.06
(1H, m, H-3′), 3.89 (1H, m, H-4′), 3.64 (2H, m, H-5′a + H-5′b),
3.59 (2H, m, NCH2), 3.11 (2H, m, SCH2), 2.52 (2H, m,
NCH2CH2), 2.05 (3H, s, SCH3), 1.69 (2H, m, SCH2CH2CH3),
0.99 (3H, t, J ) 7.4 Hz, SCH2CH2CH3).
N6-(2-Meth oxyeth yl)-2-(3,3,3-tr iflu or op r op ylth io)a d e-
n osin e (3k ). Prepared from 6 and 2-methoxyethyl bromide
1
in 51% yield: MS (FAB) m/z 454 (M+ + H+, 100%); H NMR
δ(DMSO-d6) 8.19 (1H, s, H-8), 7.80 (1H, br s, NH), 5.84 (1H,
d, J ) 6.2 Hz, H-1′), 5.40 (1H, d, J ) 6.2 Hz, OH), 5.29 (1H, d,
J ) 4.8 Hz, OH), 5.17 (1H, t, J ) 5.7 Hz, OH), 4.59 (1H, m,
H-2′), 4.17 (1H, m, H-3′), 3.99 (1H, m, H-4′), 3.60-3.49 (2H,
m, H-5′a + H-5′b), 3.45-3.38 (4H, m, NCH2 + OCH2), 3.26
(3H, s, OCH3), 3.18 (2H, m, SCH2), 2.70 (2H, m, SCH2CH2-
CF3).
2. Meth od B, Sch em e 2. 2-(3,3,3-Tr iflu or op r op ylth io)-
a d en osin e (5). A suspension of sodium hydride (1.44 g of 60%
in mineral oil, 36 mmol) and adenosine-2-thione monohydrate
(4) (5.32 g, 16.8 mmol) in dry DMF (80 mL) was stirred for 1
h before the addition of 1-chloro-3,3,3-trifluoropropane (8.72
g, 66 mmol), and the reaction was stirred for 36 h. The
disappearance of starting material was monitored by TLC
(CH2Cl2-MeOH, 9:1). The solvent was removed under reduced
pressure, and the residue was partitioned between EtOAc and
water. Drying of the organic phase and evaporation gave the
crude product which was purified by column chromatography
(CH2Cl2-MeOH, 9:1) to afford a fluffy off-white powder (4.98
g, 76%): MS (thermospray) m/z 396( M+ + H+, 100%); UV λmax
N6-(2-Meth ylth ioeth yl)-2-(3,3,3-tr iflu or opr opylth io)ade-
n osin e (3l). Prepared from 6 and 2-methylthioethyl chloride
in 36% yield: MS (FAB) m/z 470 (M+ + H+), 75 (100%); 1H
NMR δ(DMSO-d6) 8.27 (1H, s, H-8), 8.14 (1H, br s, NH), 5.82
(1H, d, J ) 6.2 Hz, H-1′), 5.44 (1H, d, J ) 6.2 Hz, OH), 5.20
(1H, d, J ) 4.9 Hz, OH), 5.08 (1H, t, J ) 5.3 Hz, OH), 4.57
(1H, m, H-2′), 4.11 (1H, m, H-3′), 3.92 (1H, m, H-4′), 3.58 (4H,
m, NCH2 + H-5′a + H-5′b), 3.27 (2H, m, SCH2), 2.72 (4H, m,
NCH2CH2 + SCH2CH2CF3), 2.09 (3H, s, SCH3).
Gen er a l Met h od for t h e P r ep a r a t ion of âγ-Dih a lo-
m eth ylen e-Su bstitu ted Tr ip h osp h a tes 10. Sch em e 3.
Phosphorus oxychloride (0.66 g, 4.3 mmol) was added to a
solution of nucleoside analogue 3 (or 5) (1.1 mmol) in triethyl
phosphate (12 mL) cooled to 0 °C. After 5 h the cold solution
was poured into ice-water (100 mL) containing sodium
bicarbonate (1.45 g, 17.25 mmol), and the mixture was stirred
for 45 min. The aqueous solution was washed with Et2O (2 ×
100 mL) and then applied to a column of Dowex 50Wx8 (H+
form). The column was washed with deionized water until the
eluate was at pH 6, and then the product was eluted with 2
M ammonium hydroxide. The product-containing fractions
were combined and freeze-dried.
The 5′-monophosphate product 9 (0.85 mmol) and tri-n-
butylamine (0.157 g, 0.85 mmol) were combined in a small
volume of pyridine and evaporated to dryness under reduced
pressure. The residue was rendered anhydrous by azeotropic
drying with pyridine (3 × 15 mL) followed by anhydrous DMF
(2 × 15 mL). It was finally dissolved in anhydrous DMF (10
mL), and carbonyldiimidazole (0.66 g, 4.1 mmol) was added.
After 4 h at room temperature methanol (0.209 g, 6.5 mmol)
was added, and 30 min later a solution of the mono(tri-n-
butylammonium) salt of the appropriate (dihalo)methylenebis-
(phosphonic acid) (4.85 mmol) in anhydrous DMF (30 mL) was
introduced. The mixture was stirred at room temperature for
18 h, then filtered, and evaporated to dryness under reduced
pressure. The crude product was purified by ion-exchange
chromatography using DEAE-Sephadex and triethylammo-
nium bicarbonate or ammonium bicarbonate solution (gradient
0-0.6 M) as eluant. The product-containing fractions were
concentrated by freeze-drying and assayed by 31P NMR. Those
found to be pure were combined and freeze-dried again to give
the product as an amine salt. If the amine salt was hygroscopic
1
(EtOH) 207 nm (ꢀ 11 200), 234 (ꢀ 23 300), 276 (ꢀ 15 000); H
NMR δ(DMSO-d6) 8.26 (1H, s, H-8), 7.46 (2H, br s, NH2), 5.82
(1H, d, J ) 6.2 Hz, H-1′), 5.43 (1H, d, J ) 6.19 Hz, OH), 5.18
(1H, d, J ) 4.4 Hz, J ) 4.4 Hz, OH), 5.04 (1H, t, J ) 5.75 Hz,
OH), 4.59 (1H, m, H-2′), 4.11 (1H, m, H-3′), 3.92 (1H, m, H-4′),
3.62-3.54 (2H, m, H-5′a + H-5′b), 3.25 (2H, m, SCH2), 2.71
(2H, m, SCH2CH2CF3).
N6,O,O,O-Tetr a a cetyl-2-(3,3,3-tr iflu or op r op ylth io)a d e-
n osin e (6). The trifluoropropylthio compound 5 (5.15 g, 13
mmol), anhydrous sodium acetate (0.723 g, 8.8 mmol), and
acetic anhydride (42 mL, 447 mmol) were heated at 80 °C for
7 h. After cooling the solution was diluted with water (100
mL), stirred at room temperature for 18 h, and then extracted
into CH2Cl2 (4 × 200 mL). The organic extracts were combined,
washed thoroughly with saturated sodium bicarbonate solution
(200 mL), dried, and evaporated. The crude product was
purified by column chromatography (Et2O-MeOH, 97:3) to
give a colorless foam (5.35 g, 73%): MS (FAB) m/z 564 (M+
+
H+), 139 (100%); H NMR δ(CDCl3) 9.36 (1H, br s, NH), 8.10
(1H, s, H-8), 6.14 (1H, d, J ) 5.3 Hz, H-1′), 5.92 (1H, t, J ) 5.5
Hz, H-2′), 5.62 (1H, t, J ) 5.5 Hz, H-3′), 4.44-4.37 (2H, m,
H-4′ + H-5′a), 4.34 (1H, m, H-5′b), 3.18 (2H, m, SCH2), 2.78
(2H, m, SCH2CH2CF3), 2.15 (3H, s, CH3CO), 2.12 (3H, s, CH3-
CO), 2.11 (3H, s, CH3CO), 2.10 (3H, s, CH3CO).
1
N6-Alk yl-2-(3,3,3-tr iflu or op r op ylth io)a d en osin es 3j-l.
The tetraacetylated compound 6 (5.067 g, 9 mmol) in dry DMF
(100 mL) was added over 3 h to a suspension of sodium hydride
(60% in mineral oil, 0.44 g, 11 mmol) in dry DMF (100 mL)
containing an appropriate alkyl iodide (27 mmol). The reaction

Antagonists of the Platelet P2T Receptor
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2 219
or deliquescent the sodium salt was prepared: The amine salt
was dissolved in methanol (2 mL) and treated with a 1 M
solution of sodium iodide in analytical grade acetone (30 mL).
The precipitate was collected by centrifugation, washed by
repeated suspension in analytical grade acetone (4 × 40 mL),
and recentrifugation. Finally the colorless solid product was
redissolved in deionized water and freeze-dried.
2-(Eth ylth io)-5′-a d en ylic Acid , Mon oa n h yd r id e w ith
Dich lor om eth ylen ebis(p h osp h on ic a cid ) (10c). Prepared
as the tetrasodium salt in 9% yield from 3c: 1H NMR δ(D2O)
8.33 (1H, s, H8), 6.15 (1H, d, J ) 5.29 Hz, H1′), 4.84 (1H, m,
H2′), 4.62 (1H, m, H3′), 4.42 (1H, m, H4′), 4.36 (2H, m, H5′a
and H5′b), 3.29 (2H, m, SCH2), 1.13 (3H, t, J ) 7.3 Hz,
SCH2CH3); 31P NMR δ(D2O) 9.23 (d, 1P, J ) 18.7 Hz, Pγ), 1.51
(dd, 1P, J 1 ) 18.7 Hz, J 2 ) 28.6 Hz, Pâ), -9.8 (d, 1P, J ) 28.7
Hz, PR). Anal. (C13H16Cl2N5Na4O12P3S‚4.5H2O) H, N, S; C:
calcd, 19.40; found, 20.13.
1H NMR δ(D2O) 8.38 (1H, s, H8), 6.13 (1H, d, J ) 5.7 Hz, H1′),
4.60 (1H, t, J ) 4.5 Hz, H3′), 4.39 (1H, m, H4′), 4.28 (2H, m,
H5′a and H5′b), 3.34 (2H, t, J ) 6.3 Hz, SCH2), 2.69 (2H, m,
SCH2CH2CF3); 31P NMR δ(D2O) 9.22 (d, 1P, J ) 19 Hz, Pγ),
1.33 (dd, 1P, J 1 ) 19 Hz, J 2 ) 29 Hz, Pâ), -9.12 (d, 1P, J )
29.0 Hz, PR). Anal. (C14H15Cl2F3N5Na4O12P3S‚3H2O) (C, H, N,
S, H2O.
N6-(2,2,2-Tr iflu or oeth yl)-2-(3,3,3-tr iflu or op r op ylth io)-
5′-a d en ylic Acid , Mon oa n h yd r id e w ith Dich lor om eth yl-
en ebis(p h osp h on ic a cid ) (10j). Prepared as the triammo-
nium salt in 4% yield from 3j: 1H NMR δ(D2O) 8.32 (1H, s,
H8), 6.60 (1H, d, J ) 5.7 Hz, H1′), 4.68 (1H, m, H2′), 4.52 (1H,
m, H3′), 4.32-4.20 (5H, m, H4′ + H5′a and H5′b + NHCH2-
CF3), 3.24 (2H, m, SCH2), 2.60 (2H, m, SCH2CH2CF3); 31P NMR
δ(D2O) 8.82 (d, 1P, J ) 18.6 Hz, Pγ), 0.63 (dd, 1P, J 1 ) 18.9
Hz, J 2 ) 28.9 Hz, Pâ), -9.43 (d, 1P, J ) 29.0 Hz, PR). Anal.
(C16H29Cl2F6N8O12P3S‚4H2O) C, H, N, H2O; S: calcd, 3.53;
found, 2.90.
2-(P r op ylth io)-5′-a d en ylic Acid , Mon oa n h yd r id e w ith
Diflu or om eth ylen ebis(p h osp h on ic a cid ) (10d ). Prepared
as a triammonium salt in 5% yield from 3d : 1H NMR δ(D2O)
8.28 (1H, s, H8), 6.01 (1H, d, J ) 4.6 Hz, H1′), 4.80 (1H, m,
H2′), 4.46 (1H, m, H3′), 4.28 (1H, m, H4′), 4.15 (2H, m, H5′a
and H5′b), 3.05 (2H, t, J ) 6.9 Hz, SCH2), 1.63 (2H, m,
SCH2CH2CH3), 0.91 (3H, t, J ) 7.3 Hz, SCH2CH2CH3); 31P
N6-(2-Met h oxyet h yl)-2-(3,3,3-t r iflu or op r op ylt h io)-5′-
a d en ylic Acid , Mon oa n h yd r id e w ith Dich lor om eth yl-
en ebis(p h osp h on ic a cid ) (10k ). Prepared as the tetrasodi-
um salt in 10% yield from 3k : 1H NMR δ(D2O) 8.36 (1H, s,
H8), 6.10 (1H, d, J ) 5.8 Hz, H1′), 4.88 (1H, m, H2′), 4.60 (1H,
m, H3′), 4.41 (1H, m, H4′), 4.38 (2H, m, H5′a and H5′b), 3.81
(2H, m, NHCH2CH2OCH3), 3.78 (2H, m, NHCH2CH2OCH3),
3.46 (3H, s, OCH3), 3.23 (2H, m, SCH2), 2.57 (2H, m, SCH2CH2-
CF3); 31P NMR δ(D2O) 9.0 (d, 1P, J ) 18.9 Hz, Pγ), 1.54 (dd,
1P, J 1 ) 18.8 Hz, J 2 ) 29.3 Hz, Pâ), -9.9 (d, 1P, J ) 29.7 Hz,
PR). Anal. (C17H21Cl2F3N5Na4O13P3S‚7H2O) C, H, N; S: calcd,
3.28; found, 2.81.
NMR δ(D2O) 4.85 (dt, 1P, J ) 53 Hz, Pγ), -1.71 (ddt, 1P, J 1
)
52 Hz, J 2 ) 28.9 Hz, Pâ), -9.44 (d, 1P, J ) 28.9 Hz, PR). Anal.
(C14H31F2N8O12P3S‚4H2O) C, H, N; S: calcd, 4.34; found, 4.89.
2-(P r op ylth io)-5′-a d en ylic Acid , Mon oa n h yd r id e w ith
Dich lor om eth ylen ebis(p h osp h on ic a cid ) (10e). Prepared
as the tetrasodium salt in 7% yield from 3d : 1H NMR δ(D2O)
8.34 (1H, s, H8), 6.11 (1H, d, J ) 5.1 Hz, H1′), 4.81 (1H, m,
H2′), 4.66 (1H, m, H3′), 4.44 (1H, m, H4′), 4.32 (2H, m, H5′a
and H5′b), 3.15 (2H, t, J ) 6.8 Hz, SCH2), 1.76 (2H, m,
SCH2CH2CH3), 1.04 (3H, t, J ) 7.3 Hz, SCH2CH2CH3); 31P
N6-(2-Meth ylth ioeth yl)-2-(3,3,3-tr iflu or op r op ylth io)-5′-
a d en ylic Acid , Mon oa n h yd r id e w ith Dich lor om eth yl-
en ebis(p h osp h on ic a cid ) (10l). Prepared as the triammo-
nium salt in 4% yield from 3l: 1H NMR δ(D2O) 8.30 (1H, s,
H8), 5.97 (1H, d, J ) 5.5 Hz, H1′), 4.65 (1H, m, H2′), 4.47 (1H,
m, H3′), 4.28 (1H, m, H4′), 4.17 (2H, m, H5′a and H5′b), 3.67
(br s, NHCH2), 3.21 (2H, t, J ) 7.6 Hz, SCH2), 2.72 (2H, t, J
) 6.6 Hz, SCH2CH2CF3), 2.58 (2H, m, NCH2CH2), 2.04 (3H, s,
SCH3); 31P NMR δ(D2O) 8.80 (d, 1P, J ) 18.6 Hz, Pγ), 0.42
(dd, 1P, J 1 ) 18.9 Hz, J 2 ) 28.9 Hz, Pâ), -9.41 (d, 1P, J ) 29.0
Hz, PR). Anal. (C17H34Cl2F3N8O12P3S2‚3H2O) H, N, S; C: calcd,
23.16; found, 23.66.
Biologica l Assa ys. The affinity of the test compounds for
the P2T receptor was assayed using washed human platelets
by the method of Humphries et al.11 Antithrombotic activity
was measured using a cyclic flow reduction model in dog
femoral artery.21 Functional kinetics of the compounds were
determined in the anesthetized rat by the method of Humphries
et al.22
δNMR (D2O) 9.15 (d, 1P, J ) 18.6 Hz, Pγ), 1.39 (dd, 1P, J 1
)
18.7 Hz, J 2 ) 28.9 Hz, Pâ), -9.6 (d, 1P, J ) 28.9 Hz, PR). Anal.
(C14H18Cl2N5Na4O12P3S‚3H2O) C, H, N, S.
N6-(2,2,2-Tr iflu or oeth yl)-2-(pr opylth io)-5′-aden ylic Acid,
Mon oa n h yd r id e w ith Dich lor om eth ylen ebis(p h osp h o-
n ic a cid ) (10f). Prepared as the tetrasodium salt in 12% yield
from 3f: 1H NMR δ(D2O) 8.25 (1H, s, H8), 5.95 (1H, d, J )
5.7 Hz, H1′), 4.84 (1H, m, H2′), 4.63 (1H, m, H2′), 4.50 (1H,
m, H3′), 4.27-4.12 (5H, m, H4′, H5′a and H5′b, NCH2), 2.97
(2H, t, J ) 6.8 Hz, SCH2), 1.58 (2H, m, SCH2CH2CH3), 0.86
(3H, t, J ) 7.35 Hz, SCH2CH2CH3); 31P NMR δ(D2O) 9.70 (d,
1P, J ) 18.5 Hz, Pγ), 3.43 (dd, 1P, J 1 ) 18.5 Hz, J 2 ) 30.38
Hz, Pâ), -9.20 (d, 1P, J ) 30.41 Hz, PR). Anal. (C16H19F3Cl2N5-
Na4O12P3S‚9H2O) C, H, N, S, H2O.
N6-(2-Meth oxyeth yl)-2-(p r op ylth io)-5′-a d en ylic Acid ,
Mon oa n h yd r id e w ith Dich lor om eth ylen ebis(p h osp h o-
n ic a cid ) (10 g). Prepared as the tetrasodium salt in 6% yield
from 3g: 1H NMR δ(D2O) 8.39 (1H, s, H8), 6.14 (1H, d, J )
5.85 Hz, H1′), 4.84 (1H, m, H2′), 4.66 (1H, m, H3′), 4.44 (1H,
m, H4′), 4.32 (2H, m, H5′a and H5′b), 3.86 (2H, m, NHCH2CH2-
OCH3), 3.80 (2H, m, NHCH2CH2OCH3), 3.46 (3H, s, OCH3),
3.22 (2H, t, J ) 7.1 Hz, SCH2), 1.80 (2H, m, SCH2CH2CH3),
1.07 (3H, t, J ) 7.35 Hz, SCH2CH2CH3); 31P NMR δ(D2O) 9.05
(d, 1P, J ) 18.7 Hz, Pγ), 1.44 (dd, 1P, J 1 ) 18.8 Hz, J 2 ) 29.3
Hz, Pâ), -9.4 (d, 1P, J ) 29.5 Hz, PR). Anal. (C17H24Cl2N5-
Na4O13P3S‚2H2O) C, N, S; H: calcd, 3.16; found, 3.82.
N6-(2-Meth ylth ioeth yl)-2-(p r op ylth io)-5′-a d en ylic Acid ,
Mon oa n h yd r id e w ith Dich lor om eth ylen ebis(p h osp h o-
n ic a cid ) (10h ). Prepared as the triammonium salt in 6% yield
from 3h : 1H NMR δ(D2O) 8.32 (1H, s, H8), 6.05 (1H, d, J )
5.5 Hz, H1′), 4.71 (1H, m, H2′), 4.52 (1H, m, H3′), 4.38 (1H,
m, H4′), 4.24 (2H, m, H5′a and H5′b), 3.67 (br s, NHCH2), 3.15
(2H, t, J ) 6.9 Hz, SCH2), 2.58 (2H, m, NCH2CH2), 2.04 (3H,
s, SCH3), 1.72 (2H, m, SCH2CH2CH3), 1.04 (3H, t, J ) 7.2 Hz,
SCH2CH2CH3); 31P NMR δ(D2O) 8.83 (d, 1P, J ) 18.6 Hz, Pγ),
0.49 (dd, 1P, J 1 ) 18.9 Hz, J 2 ) 28.9 Hz, Pâ), -9.48 (d, 1P, J
) 29.0 Hz, PR). Anal. (C17H37Cl2N8O12P3S2‚3H2O) C, H, N, S.
2-(3,3,3-Tr iflu or opr opylth io)-5′-aden ylic Acid, Mon oan -
h yd r id e w ith Dich lor om eth ylen ebis(p h osp h on ic a cid )
(10i). Prepared as the tetrasodium salt in 21% yield from 5:
Refer en ces
(1) Burnstock, G.; Kennedy, C. Is There a Basis for Distinguishing
Two Types of P2 Purinoceptor? Gen. Pharmacol. 1985, 16, 433-
440.
(2) Gordon, J . L. Extracellular ATP: Effects, Sources and Fate.
Biochem. J . 1986, 233, 309-319.
(3) Burnstock, G. P2 Purinoceptors: Historical Perspective and
Classification. In P2 Purinoceptors: Localisation, Function and
Transduction Mechanisms (Ciba Foundation Symposium 1995);
Chadwick, D. J ., Goode, J . A., Eds.; J ohn Wiley and Sons:
Chichester, 1996; pp 1-29.
(4) Daniel, J . L.; Dangelmaier, C.; J in, J .; Ashby, B.; Smith, J . B.;
Kunapuli, S. P. Molecular Basis for ADP-induced Platelet
Activation. J . Biol. Chem. 1998, 273, 2024-2029.
(5) Gerster, J . F.; J ones, J . W.; Robins, R. K. Purine Nucleosides
IV. The Synthesis of 6-Halogenated 9-â-D-Ribofuranosylpurines
from Inosine and Guanosine. J . Org. Chem. 1963, 28, 945-948.
(6) Nair, V.; Young, D. A. Photoinduced Alkylthiolation of Haloge-
nated Purine Nucleosides. Synthesis 1986, 450-453.
(7) Nair, V.; Richardson, S. G. Modification of Nucleic Acid Bases
via Radical Intermediates: Synthesis of Dihalogenated Purine
Nucleosides. Synthesis 1982, 670-672.
(8) Kikugawa, K.; Suehiro, H.; Yanase, R.; Aoki, A. Platelet Ag-
gregation Inhibitors. IX. Chemical Transformation of Adenosine
into 2-Thioadenosine Derivatives. Chem. Pharm. Bull. (Tokyo)
1977, 25, 1959-1969.
(9) Yoshikawa, M.; Kato, T.; Takenishi, T. A Novel Method for
Phosphorylation of Nucleosides to 5′-Nucleotides. Tetrahedron
Lett. 1967, 5065-5068.

220 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 2
Ingall et al.
(18) O’Connor, S. E.; Wood, B. E.; Leff, P. Characterisation of P2X
receptors in Rabbit Isolated Ear Artery. Br. J . Pharmacol. 1990,
101, 640-644.
(19) Dainty, I. A.; Leff, P.; McKechnie, K.; O’Connor, S. E. Further
Subclassification of ATP Receptors Based on Agonist Studies.
Trends Pharmacol. Sci. 1991, 12, 137-141.
(20) Nassim, M. A.; Gardner, J . J .; Wilkinson, D.; Corfield, J . E.;
Rudol, L.; Wyld, P. J . The Short-acting P2T-purinoceptor
Antagonist, FPL 67085, Reliably, Reversibly and Safely Inhibits
ADP-induced Platelet Aggregation ex vivo in Man. Br. J . Clin.
Pharmacol. 1995, 39, 98P.
(21) Humphries, R. G.; Tomlinson, W.; Leff, P.; Ingall, A. H.; Kindon,
N. D. The P2T Purinoceptor Antagonist FPL 67085, is a Potent,
Efficacious and Selective Inhibitor of Dynamic Arterial Throm-
bosis in the Pentobarbitone-anaesthetised Dog. Br. J . Pharma-
col. 1994, 114, 63P.
(10) Cramer, F. C.; Schaller, H.; Staab, H. A. Darstellung von
Imidazoliden der Phosphorsa¨ure. Chem. Ber. 1961, 94, 1612-
1621.
(11) Humphries, R. G.; Tomlinson, W.; Ingall, A. H.; Cage, P. A.; Leff,
P. FPL 66096: A Novel, Highly Potent and Selective Antagonist
at Human Platelet P2T-purinoceptors. Br. J . Pharmacol. 1994,
113, 1057-1063.
(12) Welford, L. A.; Cusack, N. J .; Hourani, S. M. O. ATP Analogues
and the Guinea Pig Tenia Coli: A Comparison of the Structure-
activity Relationships of Ectonucleotidases with those of the P2-
purinoceptor. Eur. J . Pharmacol. 1986, 129, 217-224.
(13) Blackburn, G. M.; Eckstein, F.; Kent, D. E.; Perree, T. D. Isopolar
vs Isosteric Phosphonate Analogues of Nucleotides. Nucleosides
Nucleotides 1985, 4, 165-167.
(14) Blackburn, G. M.; Taylor, G. E.; Tattershall, R. H.; Thatcher,
G. R. J .; McLennan, A. Phosphonate Analogues of Biological
Phosphates. Bioact. Mol. 1987, 3, 451-464.
-
(22) Humphries, R. G.; Tomlinson, W.; Clegg, J . A.; Ingall, A. H.;
Kindon, N. D.; Leff, P. Pharmacological Profile of the Novel P2T
purinoceptor Antagonist FPL 67085 in vitro and in the Anaes-
thetised Rat in vivo. Br. J . Pharmacol. 1995, 115, 1110-1116.
(23) Leff, P.; Robertson, M. J .; Humphries, R. G. The Role of ADP in
Thrombosis and the Therapeutic Potential of P2T-receptor An-
tagonists as Novel Antithrombotic Agents. In Purinergic Ap-
proaches in Experimental Therapeutics; J acobson, K. A., J arvis,
M. J ., Eds.; Wiley-Liss Inc.: New York, 1997; pp 203-216.
(15) Cusack, N. J .; Hourani, S. M. O.; Loizou, G. D.; Welford, L. A.
Pharmacological Effects of Isopolar Phosphonate Analogues of
ATP on P2-purinoceptors in Guinea Pig Tenia Coli and Urinary
Bladder. Br. J . Pharmacol. 1987, 90, 791-795.
(16) Kanis, J . A.; McCloskey, E. V.; Beneton, M. N. C. Clodronate
and Osteoporosis. Maturitas 1996, 23 (Suppl.), S81-S86.
(17) Gough, G.; Maguire, M. H.; Penglis, F. Analogues of Adenosine
5′-Diphosphate. New Platelet Aggregators. Influence of Purine
Ring and Phosphate Chain Substitutions on the Platelet-
aggregating Potency of Adenosine 5′-Diphosphate. Aust. Mol.
Pharmacol. 1972, 8, 170-177.
J M981072S