Structure-activity relationships of bisphosphate nucleotide derivatives as P2Y1 receptor antagonists and partial agonists

Article abstract of DOI:10.1021/jm980657j

The P2Y₁ receptor is present in the heart, in skeletal and various smooth muscles, and in platelets, where its activation is linked to aggregation. Adenosine 3',5'- and 2',5'-bisphosphates have been identified as selective antagonists at the P2Y₁ receptor (Boyer et al. Mol. Pharmacol. 1996, 50, 1323-1329) and have been modified structurally to increase receptor affinity (Camaioni et al. J. Med. Chem. 1998, 41, 183-190). We have extended the structure-activity relationships to a new series of deoxyadenosine bisphosphates with substitutions in the adenine base, ribose moiety, and phosphate groups. The activity of each analogue at P2Y₁ receptors was determined by measuring its capacity to stimulate phospholipase C in turkey erythrocyte membranes (agonist effect) and to inhibit phospholipase C stimulation elicited by 10 nM 2-(methylthio)adenosine 5'-diphosphate (antagonist effect). 2'-Deoxyadenosine bisphosphate analogues containing halo, amino, and thioether groups at the 2-position of the adenine ring were more potent P2Y₁ receptor antagonists than analogues containing various heteroatom substitutions at the 8-position. An N⁶-methyl-2-chloro analogue, 6, was a full antagonist and displayed an IC₅₀ of 206 nM. Similarly, N⁶- methyl-2-alkylthio derivatives 10, 14, and 15 were nearly full antagonists of IC₅₀ < 0.5 μM. On the ribose moiety, 2'-hydroxy, 4'-thio, carbocyclic, and six-membered anhydrohexitol ring modifications have been prepared and resulted in enhanced agonist properties. The 1,5-anhydrohexitol analogue 36 was a pure agonist with an EC₅₀ of 3 μM, i.e., similar in potency to ATP. 5'-Phosphate groups have been modified in the form of triphosphate, methyl phosphate, and cyclic 3',5'-diphosphate derivatives. The carbocyclic analogue had enhanced agonist efficacy, and the 5'-O-phosphonylmethyl modification was tolerated, suggesting that deviations from the nucleotide structure may result in improved utility as pharmacological probes. The N₆-methoxy modification eliminated receptor affinity. Pyrimidine nucleoside 3',5'- bisphosphate derivatives were inactive as agonists or antagonists at P2Y receptor subtypes.

Full text of DOI:10.1021/jm980657j

J . Med. Chem. 1999, 42, 1625-1638
1625
Str u ctu r e-Activity Rela tion sh ip s of Bisp h osp h a te Nu cleotid e Der iva tives a s
P 2Y1 Recep tor An ta gon ists a n d P a r tia l Agon ists
Erathodiyil Nandanan,† Emidio Camaioni,† Soo-Yeon J ang,† Yong-Chul Kim,† Gloria Cristalli,| Piet Herdewijn,
J ohn A. Secrist, III,§ Kamal N. Tiwari,§ Arvind Mohanram,‡ T. Kendall Harden,‡ J ose´ L. Boyer,‡ and
Kenneth A. J acobson*,†
Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes, Digestive and
Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892-0810, Department of Pharmacology, University of
North Carolina, School of Medicine, Chapel Hill, North Carolina 27599-7365, Organic Chemistry Research Department,
Southern Research Institute, P.O. Box 55305, Birmingham, Alabama 35255-5305, Departimento di Scienze Chimiche,
Universita` di Camerino, 62032 Camerino, Italy, and Laboratory of Pharmaceutical Chemistry,
Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium
Received November 23, 1998
The P2Y1 receptor is present in the heart, in skeletal and various smooth muscles, and in
platelets, where its activation is linked to aggregation. Adenosine 3′,5′- and 2′,5′-bisphosphates
have been identified as selective antagonists at the P2Y1 receptor (Boyer et al. Mol. Pharmacol.
1996, 50, 1323-1329) and have been modified structurally to increase receptor affinity
(Camaioni et al. J . Med. Chem. 1998, 41, 183-190). We have extended the structure-activity
relationships to a new series of deoxyadenosine bisphosphates with substitutions in the adenine
base, ribose moiety, and phosphate groups. The activity of each analogue at P2Y1 receptors
was determined by measuring its capacity to stimulate phospholipase C in turkey erythrocyte
membranes (agonist effect) and to inhibit phospholipase C stimulation elicited by 10 nM
2-(methylthio)adenosine 5′-diphosphate (antagonist effect). 2′-Deoxyadenosine bisphosphate
analogues containing halo, amino, and thioether groups at the 2-position of the adenine ring
were more potent P2Y1 receptor antagonists than analogues containing various heteroatom
substitutions at the 8-position. An N6-methyl-2-chloro analogue, 6, was a full antagonist and
displayed an IC50 of 206 nM. Similarly, N6-methyl-2-alkylthio derivatives 10, 14, and 15 were
nearly full antagonists of IC50 < 0.5 µM. On the ribose moiety, 2′-hydroxy, 4′-thio, carbocyclic,
and six-membered anhydrohexitol ring modifications have been prepared and resulted in
enhanced agonist properties. The 1,5-anhydrohexitol analogue 36 was a pure agonist with an
EC50 of 3 µM, i.e., similar in potency to ATP. 5′-Phosphate groups have been modified in the
form of triphosphate, methyl phosphate, and cyclic 3′,5′-diphosphate derivatives. The carbocyclic
analogue had enhanced agonist efficacy, and the 5′-O-phosphonylmethyl modification was
tolerated, suggesting that deviations from the nucleotide structure may result in improved
utility as pharmacological probes. The N6-methoxy modification eliminated receptor affinity.
Pyrimidine nucleoside 3′,5′-bisphosphate derivatives were inactive as agonists or antagonists
at P2Y receptor subtypes.
In tr od u ction
P2Y6, at which uracil nucleotides are particularly active,
and P2Y11, which is activated by ATP.7,8 Among the
three uracil nucleotide-responsive P2Y receptors, the
P2Y2 is activated equally by ATP and UTP. While the
P2Y4 subtype is mainly activated by UTP, the P2Y6 is
activated by UDP. Recently it was found that uracil
nucleotides are released from cells in a manner similar
to adenine nucleotides, and this has implications for
P2Y receptor activation.9
Extracellular adenine and uracil nucleotides act in
cell signaling through two families of plasma membrane
receptors: the G protein-coupled receptors and the
ligand-gated cation channels, termed P2Y and P2X,
respectively.1,2 Molecular biological techniques have led
to the discovery of new nucleotide receptor subtypes.
The P2Y1 receptor for purine nucleotides from chick
brain was the first subtype of the P2Y receptor to be
cloned.3 The human P2Y1 subtype4,5 has also been
cloned, and the gene for the human P2Y1 receptors is
localized to chromosome 3.6 Four other distinct, cloned
P2Y subtypes are P2Y2 (formerly called P2U), P2Y4,
Agonist binding at P2Y1 receptors results in activation
of phospholipase C (PLC), which generates inositol
phosphates and diacylglycerol from phosphatidylinositol
4,5-bisphosphate.10 The P2Y1 receptor is present in the
heart, in skeletal and various smooth muscles, and in
the brain. A P2Y1 receptor in platelets is involved in
ADP-promoted aggregation.11,12 Thus, a selective P2Y1
receptor antagonist may have potential as an anti-
thrombotic agent. A selective P2Y1 receptor agonist also
has potential as an antihypertensive and possibly
antidiabetic agent.13,14
* Correspondence to: Dr. Kenneth A. J acobson, Chief, Molecular
Recognition Section, Bldg 8A, Rm B1A-19, NIH, NIDDK, LBC,
Bethesda, MD 20892-0810. Tel: (301) 496-9024. Fax: (301) 480-8422.
E-mail: kajacobs@helix.nih.gov.
† NIH.
‡ University of North Carolina, School of Medicine.
§ Southern Research Institute.
| Universita` di Camerino.
Katholieke Universiteit Leuven.
10.1021/jm980657j CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/21/1999

1626 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
Ta ble 1. Structures of Deoxyadenosine 3′,5′-Bisphosphate
various heteroatom substitutions at the 8-position (Table
1). Carbocyclic, 4′-thio, and six-membered ring modifica-
tions have been prepared on the ribose moiety. 5′-
Phosphate groups have been modified in the form of
triphosphate, methylphosphonate, and cyclic 3′,5′-
diphosphate derivatives (Figure 1). Pyrimidine 3′,5′-
bisphosphate analogues also were included (Figure 1),
since other subtypes of P2Y receptors are known to
recognize uridine 5′-di- and triphosphates.1,7
Derivativesa
Some of the nucleosides utilized for phosphorylation
were obtained commercially, leading to compounds 3,
5, 17b, and 32-35. Other nucleosides were synthesized
as reported, leading to compounds 4, 18, 19, 23, 24, 30,
31, and 36,24-28 or alternatively were synthesized as
shown in Schemes 1-6.
compd
R
R′
R′′
X
Y
Z
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17a
17b
18
19
20
21
22
23
24
25
NH2
NH-CH3
NH2
NH-CH3
NH2
NH-CH3
NH-CH3
NH-OCH3
NH2
NH-CH3
NH2
NH-CH3
NH2
NH-CH3
NH-CH3
NH-CH3
NH-CH3
OH
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NHCH3
H
H
H
H
Cl
Cl
I
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
N
H
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
N
N
CH
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2-Halo groups were introduced in the structure of the
lead compound 2 resulting in compounds 6 (Scheme 1)
and 7 (Scheme 2). Commercially available cladribine,
37 (2-chloro-2′-deoxyadenosine), was acetylated29 to give
the diacetate 38 which upon diazotization-iodination
followed by treatment with methylamine resulted in the
N6-methyl derivative 39 in 42% yield. Compound 39 was
phosphorylated to give the 2-chloro analogue 6. The
2-iodo analogue 7 was prepared in seven steps (Scheme
2). 2′-Deoxyguanosine, 45, was acetylated to the diac-
etate 46 in 75% yield using acetic anhydride and
pyridine, as reported in the literature.29 Chlorination
of 46 with POCl3 in the presence of N,N-dimethylaniline
and tetraethylammonium chloride in acetonitrile at 100
°C afforded the 6-chloro derivative 47, which on treat-
ment with methylamine resulted in 2-amino-N6-methyl-
adenosine, 48, in 55% yield. Acetylation29 of 48 followed
by diazotization-iodination and hydrolysis furnished
2-iodo-N6-methyladenosine, 50, in 48% yield (Scheme
2).
S-CH3
S-CH3
S-CH2CH3
S-CH2CH3
S-(CH2)2CH3
S-(CH2)2CH3
S-CH2CHdCH2
S-(CH2)5CH3
NH2
NH2
H
H
H
H
H
H
H
CH3
CHdCH2
O-CH3
S-CH3
NH-CH3
H
S
H
H
CH2
CH2
NH2
a
The structures of compounds 26-36 are given in Figure 2.
2′-Deoxyadenosine nucleotides modified at the 2-posi-
tion with thioether (9-16) and amino (17) groups were
also synthesized. Cladribine, 37, on nucleophilic dis-
placement with the potassium salt of methyl-, ethyl-,
or propylthiol afforded the corresponding 2-thio deriva-
tives 51-53, in 70-80% yield (Scheme 3), leading to
phosphorylated 2-alkylthio analogues 9, 11, and 13. To
obtain the corresponding N6-methyl derivatives, 2-chloro-
N6-methyl-2′-deoxyadenosine (39; (Scheme 1) was con-
verted to various 2-thioethers by displacement of chlo-
rine at the 2-position with the corresponding alkanethiol
potassium salts in 70-80% yield, leading to compounds
10, 12, and 14-16.
The synthetic approach for the modification at the
8-position utilized commercially available 2′-deoxyad-
enosine, 54, as the starting compound (Scheme 4).
Bromination of 54 using NBS yielded the 8-bromo
derivative, 55, in 40% yield. Displacement of bromine
on 53 with various nucleophiles, sodium methoxide,
methylamine, or sodium thiomethoxide furnished the
corresponding substituted products 56-58 in ∼70-90%
yield. Phosphorylation of 56-58 and 8-methyl- or
8-vinyl-substituted 2′-deoxyadenosine derivatives30 us-
ing the general phosphorylation conditions resulted in
the corresponding bisphosphates 18-22 in low to mod-
erate yields.
Few potent and selective P2 receptor antagonists are
known.15-18 However, adenosine 3′,5′- and 2′,5′-bispho-
sphates were recently shown to be selective inhibitors
of P2Y1 receptors,19 and a novel series of bisphosphate
adenosine analogues has been synthesized as potent and
selective P2Y1 receptor antagonists.20 Among them, N6-
methyl-2′-deoxyadenosine 3′,5′-bisphosphate (MRS 2179,
2; Table 1) was synthesized and is a competitive
antagonist at the turkey P2Y1 receptor with a KB value
of 102 nM.21 MRS 2179 was inactive at P2Y2, P2Y4, and
P2Y6 subtypes, at the adenylyl cyclase-linked P2Y
receptor in C6 glioma cells (J . Boyer, personal com-
munication),22 and at a recently cloned, novel avian P2Y
receptor that inhibits adenylyl cyclase.23 The antagonist
MRS 2179, although competitive and selective, does not
have sufficiently high affinity for use as a radioligand.
Thus, further structural modifications directed at in-
creasing affinity as well as biological stability have been
carried out in the present study.
Resu lts
Ch em ica l Syn th esis. The recent discovery of N6-
methyl-2′-deoxyadenosine 3′,5′-bisphosphate (2; Table
1) as a competitive antagonist at P2Y1 receptors20,21 led
us to explore further structural modifications of this
molecule in order to increase P2Y1 receptor affinity and
selectivity. We have prepared 2′-deoxyadenosine bis-
phosphate analogues containing halo, amino, and thio-
ether groups at the 2-position of the adenine ring or
Compound 8, i.e., the methoxy analogue of 2, was
prepared. The nucleoside precursor to be phosphory-
lated, 59, was synthesized by reaction of methoxyamine

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1627
F igu r e 1. Structures of purine and pyrimidine nucleotide analogues. The corresponding ammonium salts were synthesized and
tested for biological activity.
Sch em e 1. Synthesis of Nucleoside Precursors of
2-Chloro- and 2-Alkylthio-Substituted Analogues of
N6-Methyladenosine 3′,5′-Bisphosphatea
Sch em e 2. Synthesis of 2-Iodo-N6-methyladenosine and
Related Nucleosidesa
a
(a)Ac2O, pyridine, 25 °C, 2 h; (b) (1) IAN, CH2I2, 85 °C, 1 h,
(2) 40% aq MeNH2, 25 °C, 1 h; (c) RSH, KButO, or CH3SNa, DMF,
110 °C, 4 h; (d) (1) POCl3, (MeO)3PO, Proton Sponge, 0 °C, 1 h, (2)
NH4HCO3.
a
(a) Ac2O, Py, 65 °C, 4 h; (b) POCl3, NNDA, TEA Cl, CH3CN,
90 °C, 10 min; (c) 40% aq MeNH2, 25 °C, 2 h; (d) Ac2O, Py, 25 °C,
4 h; (e) (1) POCl3, (MeO)3PO, Proton Sponge, 0 °C, 1 h, (2)
NH4HCO3; (f) IAN, CH2I2, 85 °C, 1 h.
and 6-chloro-2′-deoxypurine riboside using the general
method previously described.20
We prepared the bisphosphates of 1,5-anhydro-2-
(adenin-9-yl)-2,3-dideoxy-D-arabino-hexitol,25 4′-thioad-
enosine,26,31 23, and carbocyclic32 analogues, 24 and 25,
of 2′-deoxyadenosine 3′,5′-bisphosphate (Table 1). Al-
though difficulties in phosphorylation of 4′-thiopyrim-
idines were previously reported,37 we have isolated the
bisphosphate derivative of 4′-thioadenosine by HPLC of
a mixture in low yield. The optimal reaction conditions
were a brief reaction time (20 min) at 0 °C. Carbocyclic
2-amino-6-chloropurine 2′-deoxyriboside27 was phospho-
rylated to provide 24. For inclusion of 2-position sub-
stitution, carbocyclic 2-amino-6-chloropurine 2′-deoxyribo-
side was treated with methylamine and phosphorylated
(Scheme 5). 2′-Deoxy-4′-thioadenosine, 62, was also
oxidized to the corresponding sulfoxide 63, which upon
phosphorylation provided only the 5′-monophosphate 26
(Scheme 5).

1628 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
Sch em e 5. Synthesis of Precursors of Various
Sch em e 3. Synthesis of Nucleoside Precursors of
2-Alkylthio-Substituted Analogues of Adenosine
3′,5′-Bisphosphatea
Ribose-Modified Analogues (carbocyclic and 4′-thio) of
Adenosine 3′,5′-Bisphosphatea
a
(a) RSH, KButO, or CH3SNa, DMF, 110 °C, 4 h; (b) (1) POCl3,
(CH3O)3PO, Proton Sponge, 0 °C, 1 h, (2) NH4HCO3.
Sch em e 4. Synthesis of Nucleoside Precursors of
8-Substituted Analogues of Adenosine
3′,5′-Bisphosphatea
a
(a) 40% aq CH3NH2, 25 °C, 3 h; (b) (1) POCl3, (MeO)3PO,
Proton Sponge, 0 °C, 1 h, (2) NH4HCO3; (c) NaIO4, 25 °C, 1 h.
Sch em e 6. Synthesis of a 5′-O-Phosphonylmethyl
Analogue of Adenosine 3′,5′-Bisphosphatea
a
(a) NBS, DMF, 25 °C, 12 h; (b) NaOCH3, DMF, 110 °C, 6 h, or
NaSCH3, DMF 110 °C, 4 h, or MeNH2, MeOH, 25 °C, 3 h; (c) (1)
POCl3, (CH3O)3PO, Proton Sponge, 0 °C, 1 h, (2) NH4HCO3.
The effect of various substitutions and hence various
charges on the phosphates also was examined. An
O-phosphonylmethyl group was introduced selectively
at the 5′-position followed by phosphorylation at the 3′-
position (Scheme 6) resulting in compound 27. Reaction
of 2′-deoxyadenosine, 54, with diethyl tosyl methylphos-
phonate in the presence of potassium tert-butoxide in
DMF for 3 days yielded the 5′-diethyl phosphonate of
2′-deoxyadenosine 64 in 32% yield, which was phospho-
rylated by a general procedure to provide compound 27.
Alternatively, phosphorylation of 54 using the general
procedure with methyl dichlorophosphate yielded the
corresponding bis(methyl phosphate) 28 in good yield.
Attempted phosphorylation of the 3′,5′-bisphosphate 2
to prepare corresponding 5′-di- and triphosphates re-
sulted instead in isolation of 3′,5′-cyclic di- and tetra-
phosphates 29a and 29b (Scheme 7). The structures of
the 3′,5′-cyclic phosphate derivatives were assigned by
2D phosphorus and proton-phosphorus decoupling
NMR analysis.
a
(a) KButO, DMF, 25 °C, 3 days; (b) (1) POCl3, (CH3O)3PO,
Proton Sponge, 0 °C, 1 h, (2) NH4HCO3.
tained for each compound alone and in combination with
2-(methylthio)adenosine 5′-diphosphate (2-MeSADP),
which itself resulted in a marked and concentration-
dependent activation of the turkey erythrocyte phos-
pholipase C.21 2-MeSADP was used at a concentration
of 10 nM (approximately ) EC50). The activities of all
the newly synthesized analogues were compared to that
of 2, which we previously identified as the highest
potency P2Y1 receptor antagonist thus far identified.20,21
Concentration-response curves for representative com-
pounds are shown in Figure 2.
Biologica l Activity. The deoxyadenosine bisphos-
phate nucleotide analogues prepared in the present
study were tested separately for agonist and antagonist
activity in the PLC assay at the P2Y1 receptor in turkey
erythrocyte membranes,21,39 and the results are reported
in Table 2. Concentration-response curves were ob-
The presence of the 2′-hydroxyl group in the arabinose
derivative 3 compared to the corresponding 2′-deoxy
compound 1 resulted in a 3-fold loss of antagonist
potency and an increase in the agonist potency and
efficacy. The 1-deaza modification in 4 was tolerated at
P2Y1 receptors, resulting in a 3-fold loss of antagonist

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1629
cious in receptor activation. The N6-methyl-2-amino
carbocyclic analogue 25 was also a mixed agonist/
antagonist but was 2-fold less potent than 24 as an
antagonist. Compound 26, a sulfoxide 5-monophosphate,
was inactive at P2Y1 receptors.
Sch em e 7. Synthesis of 3′,5′-Cyclic Di- and
Tetraphosphate Derivativesa
Phosphate group modifications were made. The 5′-
phosphonate analogue 27 was a mixed agonist/antago-
nist. While producing a 78% maximal inhibition, 27 had
antagonist potency that was only 3.5-fold lower than for
the corresponding 5′-phosphate 1. The dimethyl phos-
phate 28 and the 3′,5′-cyclic diphosphate 29a were
essentially inactive at P2Y1 receptors. The 3′,5′-cyclic
tetraphosphate 29b displayed potent antagonist activ-
ity; however the possibility of hydrolysis of the unusual
cyclic structure during the assay was not explored. The
effect of lengthening the 5′-substituent from a mono-
phosphate to a triphosphate was examined in 3′-deoxy-
3′-azidoxylose derivatives, compounds 30 and 31. The
triphosphate was a weak agonist, while the monophos-
phate was nearly inactive, suggesting that the triphos-
phate group may tend to enhance agonist potency versus
the 5′-monophosphate, as was observed for nucleotides
phosphorylated solely at the 5′-position.33
Introduction of a variety of pyrimidine bases in
compounds 32-35 completely abolished both agonist
and antagonist activity at P2Y1 receptors. In addition,
the 2′-deoxyuridine derivative 32 at 100 µM was shown
to be inactive as either an agonist or antagonist at P2Y2,
P2Y4, or P2Y6 receptors (all of which are activated by
uracil nucleotides) and at the C6 glioma P2 receptor22
linked to inhibition of adenylyl cyclase.
a
(a) (1) Carbonyldiimidazole, DMF, 25 °C, 12 h, (2) bis(tribu-
tylammonium)pyrophosphate, 25 °C, 42 h, (3) NH4HCO3.
potency versus the corresponding aza analogue 2. The
combination of 2-halo and N6-methyl groups resulted
in higher antagonist potency for the chloro derivative 6
(IC50 0.20 µM, Figure 2) and a lower potency for the iodo
derivative 7 in comparison with 2. Both 6 and 7 were
nearly pure antagonists. The 6-(methoxyamino)purine
analogue 8 was nearly inactive at P2Y1 receptors, which
is consistent with the previously observed critical
requirement for a 6-amino or small alkylamino group.
Substitution at the adenine 2-position was generally
well-tolerated at P2Y1 receptors and was compatible
with the antagonist potency-enhancing effects of an N6-
methyl group. For example, the N6-methyl-2-methylthio
analogue 10 was nearly a full antagonist and equipotent
to the corresponding 2-H analogue 2. Compound 10 was
5-fold more potent as an antagonist than the corre-
sponding 6-NH2 analogue 9 (Figure 2), which also
displayed agonist activity. The 2-ethylthio substitution
curiously resulted in ∼3-fold lower potency than 2-me-
thylthio, either as an agonist (in the case of the
6-aminopurine, 11) or as an antagonist (with N6-methyl
substitution, 12). The N6-methyl-2-propylthio analogue
14 was an antagonist, while the corresponding 6-NH2
analogue 13 was a mixed agonist/antagonist of equal
proportions. The 2-allylthio and 2-n-hexylthio analogues
containing the N6-methyl substitution, 15 and 16,
respectively, were full antagonists. The allylthio and
n-propylthio analogues were equipotent (0.4-0.5 µM
IC50), while the n-hexylthio analogue was 7-fold less
potent as an antagonist. Thus, as for P2Y1 receptor
agonists,33,34 the 2-position can accommodate steric bulk
without substantial loss of potency as a P2Y1 receptor
antagonist. Substitutions at the 2-position other than
thioethers were also studied. The 2-amino-N6-methyl
analogue 17a was an antagonist of moderate potency.
The corresponding 6-hydroxyl (deoxyguanosine) ana-
logue 17b was an 8-fold weaker antagonist than 17a .
8-Position substitution in the methyl, vinyl, methoxy,
methylthio, and methylamino analogues 18-22 resulted
in only weak P2Y1 receptor antagonist properties,
similar to the activity previously reported for the
8-bromo analogue. Within this group, carbon substitu-
ents, such as the 8-vinyl group, in 19, were best
tolerated at P2Y1 receptors, while the 8-methyloxy
substituent, in 20, resulted in a complete loss of activity.
Ribose ring substitutions tended to enhance the
agonist properties of the derivatives. The 4′-thio ana-
logue 23 and the carbocyclic analogue 24 were mixed
agonist/antagonists, with the 4′-thio being more effica-
The ribose was also replaced with a six-membered
ring in which the glycosidic bond was absent. Thus, the
enlargement of the ribose ring in the anhydrohexitol-
adenine25 derivative 36 resulted in a pure P2Y1 receptor
agonist activity, with an EC50 of 3 µM (Figure 2).
Discu ssion
The aim of the present study was to synthesize novel
analogues and isosteres of 2 in order to increase the
affinity and/or alter the efficacy of these nucleotides as
antagonists, mixed agonists, or pure agonists at turkey
P2Y1 receptors. The major conclusions are that 2-posi-
tion substituents are compatible with the 2′-deoxy
modification and in fact may enhance affinity, as in the
2-chloro analogue 6, and that substantial modification
of the ribose moiety is possible. However, the agonist
component of the biological activity of the bisphosphate
analogues tends to increase upon 2-alkylthio substitu-
tion. This was particularly pronounced in the absence
of the N6-methyl group. The effect of length of 2-alkyl-
thio ethers on antagonist activity does not strictly follow
the trend previously observed for potency of 5′-mono-
phosphate agonists, e.g., 2-hexylthio > 2-methylthio. In
the present series the most potent antagonists among
the 2-thioethers were the S-methyl, propyl, and allyl
analogues 10, 14, and 15, respectively, which also
contained an N6-methyl group. In the 2-alkylthioether
series, the presence of the N6-methyl group greatly
decreased agonist efficacy. Amino substitution at the
2-position also was found to be tolerated in P2Y1
receptor antagonists.
Among intended isosteres 2, we observed that the
carbocyclic analogue 24 was active as a mixed agonist/

1630 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
Ta ble 2. In Vitro Pharmacological Data for Stimulation of PLC at Turkey Erythrocyte P2Y1 Receptors (agonist effect) and Inhibition
of PLC Stimulation Elicited by 10 NM 2-MeSADP (antagonist effect), for at Least Three Separate Determinations
agonist effect, %
antagonist effect, %
compd
of maximal increasea
EC50, µMa
6.29 ( 2.54
of maximal inhibitionb
IC50, µMb (n)
1
2e
3
12 ( 3
NE
44 ( 5
NE
19 ( 3
NE
13 ( 3 (n ) 2)
NE
87 ( 4
99 ( 1
56 ( 6
96 ( 2
80 ( 3
95 ( 1
87 ( 3
low potency/NE
78 ( 2
94 ( 2
84 ( 1
94 ( 1
46 (7
94 ( 3
97 ( 3
100
96 ( 2
93 ( 4
53 ( 10
59 ( 10
NE
5.76 ( 0.68
0.331 ( 0.059 (5)
14.9 ( 1.0 (4)
0.904 ( 0.239
2.01 ( 0.83
0.206 ( 0.053
0.891 ( 0.233
>100 (4)
1.89 ( 1.07
0.362 ( 0.119
4.59 ( 1.61
0.98 ( 0.27
c (4)
0.449 ( 0.190
0.475 ( 0.206
3.37 ( 0.91
1.85 ( 0.74
14.9 ( 5.1
75.1 ( 14.5
30.7 ( 14.8
2.34 ( 0.50
0.651 ( 0.160
4.44
4
5
6e
7
8
9
22 ( 2
0.550 ( 0.117
d
1.98 ( 0.42
10
11
12
13e
14
15
16
17a
17b
18
19
20
21
22
23
24
25
26
27
28
29a
29b
30
31
32
33
34
35
36e
6 ( 2
16 ( 1
NE
53 ( 7
1.71 ( 0.78
d
7 ( 2
NE
NE
4
NE
NE
NE
NE
d
NE
NE
low potency/NE
low potency/NE
38
73 ( 11
74 ( 3
NE
(2)
c
56 ( 7
12.2 ( 3.9
7.21 ( 4.40
6.49 ( 2.76
27 ( 11
2.53 ( 0.57
5.42 ( 2.13
26 ( 3
NE
22 ( 5
NE
NE
14 ( 2
46 ( 10 at 100 µM
49 ( 7 at 100 µM
NE
NE
NE
NE
100
27.7 ( 16.6
78 ( 5
NE
19.6 ( 6.1 (4)
0.158 ( 0.064
low potency/NE
87 ( 4
NE
low potency/NE
NE
NE
NE
NE
NE
0.02
low-potency agonist
low-potency agonist
2.99 ( 0.35
(2)
a
Agonist potencies were calculated using a four-parameter logistic equation and the GraphPad softaware package (GraphPad, San
Diego, CA). EC50 values (mean ( standard error) represent the concentration at which 50% of the maximal effect is achieved. Relative
efficacies (%) were determined by comparison with the effect produced by a maximal effective concentration of 2-MeSADP in the same
b
experiment. Small increase refers to <10% at 100 µM. Antagonist IC50 values (mean ( standard error) represent the concentration
needed to inhibit by 50% the effect elicited by 10 nM 2-MeSADP. The percent of maximal inhibition is equal to 100 minus the residual
fraction of stimulation at the highest antagonist concentration; n ) 3, unless otherwise indicated in parentheses. c The IC50 of partial
agonists with high intrinsic agonist efficacy (>60% of the maximal response) cannot be accurately estimated under these assay conditions
due to the small difference between the submaximal response of the full agonist (2-MeSADP) and that produced by the partial agonist.
d
EC50 was not calculated for increases of )10% at 100 µM. e 2, MRS 2179; 6, MRS 2216; 13, MRS 2242; 36, MRS 2255. NE, no effect at
100 µM. Chemical structures are given in Table 1 and Figure 1.
antagonist at P2Y1 receptors. The carbocyclic modifica-
tion is expected to increase the in vivo stability of the
compounds since these are N-cycloalkyl analogues, and
the equivalent of deglycosylation is not expected to
readily occur. The carbocyclic modification also may
increase the selectivity of the compounds for P2 recep-
tors, since these are structurally distinct from natural
nucleotides.
The balance of agonist versus antagonist properties
was highly dependent on ribose modifications, particu-
larly at the 2′- and 4′-positions. Substitution at the
ribose 2′-position generally appeared to increase the
degree of agonism observed, as is evident from the
present and previous20 sets of analogues (Figure 3). The
presence of a 2′-hydroxyl, in either configuration, or a
2′-methoxy group increased the agonist efficacy. Per-
haps the most unexpected result was that an expansion
of the ribose ring in the 1,5-anhydrohexitol-adenine
derivative, 36, results in pure agonism at the turkey
P2Y1 receptor. As for the five-membered carbocyclic
analogue 24, compound 36 would be resistant to deg-
lycosylation by the usual mechanisms.
In general, for pharmacological studies of receptor
subtypes, highly selective antagonists are preferable.
However, for some therapeutic applications partial
agonists may provide low intrinsic activities that might
avoid side effects.35 For example, pindolol is such a
partial agonist at â1-adrenergic receptors and displays
diminished bradycardiac side effects.36 Thus, the partial
agonists in this series of nucleotide derivatives may
potentially be therapeutically useful in modulating the
activity of P2Y1 receptors. The effects of partial P2Y1
receptor agonists on platelet aggregation or on insulin
release have not yet been explored.
A preliminary report37 indicates that 2 is a relatively
potent antagonist at rat P2X1 receptors with an IC50
value of 1.2 µM, as determined in a study of ATP-
induced ion flux in Xenopus oocytes transfected with the
recombinant rat P2X1 receptor. In similar experiments
using rat P2X2 and P2X4 receptors, no antagonist

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1631
more, in platelets, the presence of P2X1 antagonist
properties, as has been demonstrated for 2,37 may
increase the potential utility as antiaggregatory agents
through action at two receptors having the same net
pharmacological effect, i.e. platelet aggregation. Other
members of the present series may also prove to be dual
antagonists.
We have used mutagenesis and molecular modeling
of human P2Y1 receptors to define a hypothetical
nucleotide binding site, defined by TM3, TM6, and TM7
domains, for both the agonist 2-MeSADP and the
antagonist 2. In the present study we have determined
structural features which dramatically influence the
degree of efficacy of the compound, such that nearly
pure agonists and pure antagonists are present among
closely related derivatives.
The pyrimidine nucleotide, 2′-deoxyuridine 3′,5′-bis-
phosphate, 32, was found to be inactive at subtypes of
P2Y receptors known to be activated by uracil nucle-
otides, e.g., P2Y2, P2Y4, and P2Y6. At the P2Y2 receptor
this was not surprising, since in the SAR of nucleotide
agonists at this subtype a 5′-triphosphate group is
required for activation. However, even P2Y6 receptors,
which are activated by UDP, would not accept as ligand
the 2′-deoxy-3′,5′-bisphosphate analogue 32. Thus, the
search for effective antagonists at these subtypes con-
tinues.
F igu r e 2. Effects of deoxyadenosine bisphosphate derivatives
on phospholipase C in turkey erythrocyte membranes: both
concentration-dependent stimulation of inositol phosphate
formation by 2-MeSADP (O), compound 9 (0), and compound
36 (b) and its inhibition in the presence of 10 nM 2-MeSADP
by compound 6 (2) and compound 9 (9). Membranes from [3H]-
inositol-labeled erythrocytes were incubated for 5 min at 30
°C in the presence of the indicated concentrations of 2-Me-
SADP or of test compound, either alone or in combination with
10 nM 2-MeSADP. The data shown are typical curves for at
least three experiments carried out in duplicate using different
membrane preparations.
In conclusion, the present study has identified new
pharmacological probes of P2Y1 receptors. The SAR of
2 indicates that 2-position substitution is favorable for
affinity at P2Y1 receptors, and nearly pure antagonism
is maintained provided that the N6-methyl group is
present. An N6-methyl-2-chloro analogue 6 was a full
antagonist and displayed an IC50 of 206 nM. Similarly,
N6-methyl-2-alkylthio derivatives 10, 14, and 15 were
nearly full antagonists of IC50 < 0.5 µM. Thus, the
affinity of 2-substituted N6-methyl analogues decreases
in the order: Cl > H, SCH3 > S-propyl, S-allyl > NH2,
S(CH2)5CH3. The 2-(propylthio)-6-aminopurine analogue
13 is the most potent mixed agonist/antagonist, in which
the two components are roughly equal. Among nucle-
otides with mixed agonist and antagonist properties, but
in which the antagonist component predominates, com-
pounds 5 and 9, 2-chloro- and 2-(methylthio)-6-ami-
nopurine analogues,20 respectively, are the most potent.
The 1,5-anhydrohexitol-adenine derivative 36 is a pure
agonist of potency similar to ATP.33
F igu r e 3. Effects of 2′-position substitution of adenosine 3′,5′-
bisphosphate derivatives17 on agonist versus antagonist prop-
erties at turkey erythrocyte P2Y1 receptors. IC50 and EC50
values are given in µM.
activity was observed, and at rat P2X3 receptors only
weak antagonism was observed. The finding of affinity
of 2, an N6-methyl derivative, at P2X1 receptors is in
contrast with findings previously observed for adenosine
5′-triphosphate derivatives as agonists,33,34 in which the
presence of the N6-methyl group precluded activity at
guinea pig vas deferens or urinary bladder P2X recep-
tors. Thus, 2 is a selective antagonist for P2Y1 receptors
within the family of metabotropic P2 receptors and a
selective antagonist for P2X1 receptors within the family
of ionotropic P2 receptors. The effects on activity at P2X
receptors of the structural modifications of 2 reported
in the present study have yet to be determined.
Exp er im en ta l Section
Ch em ica l Syn th esis. Nucleosides and synthetic reagents
were purchased from Sigma Chemical Co. (St. Louis, MO) and
Aldrich (St. Louis, MO). 6-Chloro-2′-deoxypurine riboside was
obtained from Sigma. MRS 2179 was synthesized in our
laboratory as described.20 Several 2′-deoxynucleosides, includ-
ing 1-deaza-2′-deoxyadenosine,24 anhydrohexitol-adenine nu-
cleoside,25 2′-deoxy-4′-thioadenosine,26 2′-deoxyaristeromycin,27
8-alkyl-2′-deoxyadenosine,30 and 3′-deoxy-3′-azidoxylose,28 were
synthesized as reported.
In platelets, three subtypes11 of P2 receptors are
present: P2X1, P2Y1, and a yet uncloned P2T receptor,
which inhibits adenylate cyclase. A P2Y1 receptor
antagonist may prove to inhibit platelet aggregation,12
as do adenosine 5′-triphosphate analogues. There has
been some discussion of whether ATP acts as a pure
antagonist or as a partial agonist in platelets. It has
been proposed that ATP itself has lower efficacy than
ADP at a platelet P2Y receptor, and that may be why
it apparently antagonizes the effects of ADP.15 Further-
1H NMR spectra were obtained with a Varian Gemini-300
spectrometer using D2O as a solvent. 31P NMR spectra were
recorded at room temperature by use of a Varian XL-300
spectrometer (121.42 MHz); orthophosphoric acid (85%) was
used as an external standard.
Purity of compounds was checked by a Hewlett-Packard
1090 HPLC apparatus using SMT OD-5-60 RP-C18 (250 × 4.6
mm; Separation Methods Technologies, Inc., Newark, DE) as
an analytical column in two solvent systems. The flow rate

1632 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
Ta ble 3. Synthetic Data for Nucleotide Derivatives, Including Structural Verification Using HRMS and Purity Verification Using
HPLC
FAB (M - H+)
HPLC tR, mina
compd
formula
calcd
found
system A
system B
yields, %b
1
2
3
4
5
6
7
8
9
C10H15O9N5P2
410.0267
424.0423
426.0216
423.0471
443.9877
458.0034
549.9390
440.0372
456.0144
470.0301
470.0301
484.0457
484.0457
498.0614
496.0457
540.1083
439.0532
426.0216
424.0423
436.0423
440.0372
456.0144
439.0532
426.0038
408.0474
437.0740
362.0324
424.0423
438.0580
406.0318
565.9644
451.0281
610.9607
386.9995
401.0151
404.9889
386.0154
424.0423
410.0249
424.0404
426.0226
423.0479
443.9872
458.0035
549.9396
440.0358
456.0122
470.0307
470.0307
484.0453
484.0452
498.0605
496.0445
540.1057
439.0535
426.0216
424.0412
436.0429
440.0380
456.0160
439.0497
426.0051
408.0474
437.0759
362.0319
424.0432
438.0580
406.0336
565.9664
451.0288
610.9602
386.9977
401.0160
404.9901
386.0136
424.0423
5.98
6.93
6.12
8.49
7.20
8.51
7.84
6.32
8.14
9.99
8.94
10.49
7.77
9.75
9.20
12.87
6.48
9.28
6.07
6.43
4.14
6.89
6.67
6.75
5.97
3.70
2.76
7.04
3.06
7.60
6.91
3.70
3.81
4.20
4.63
3.66
2.10
4.31
11.51
12.32
10.51
10.64
12.14
11.42
8.40
10.13
12.98
12.07
9.84
11.81
11.65
12.40
12.35
15.11
7.15
9.69
8.72
9.00
7.28
8.23
9.20
8.92
7.04
5.21
7.63
7.37
4.48
7.47
12.11
5.78
5.99
9.85
9.73
7.83
3.34
7.56
44
41
15
19
58
17
16
23
48
17
23
23
21
10
13
25
19
39
9
13
17
37
5
13
24
10
2
10
22
20
5
13
3
39
39
27
36
1
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
11H17O9N5P2
10H15O10N5P2
12H18O9N4P2
10H14O9N5P2Cl
11H16O9N5P2Cl
11H16O9N5P2I
11H17O10N5P2
11H17O9N5P2S
12H19O9N5P2S
12H19O9N5P2S
13H21O9N5P2S
13H21O9N5P2S
14H23O9N5P2S
14H21O9N5P2S
17H29O9N5P2S
11H18O9N6P2
10H15O10N5P2
11H17O9N5P2
12H17O9N5P2
11H17O10N5P2
11H17O9N5P2S
11H18O9N6P2
10H15O8N5P2S
11H17O8N5P2
12H20O8N6P2
10H14O6N5PS
11H17O9N5P2
12H19O9N5P2
11H15O8N5P2
11H17O14N5P4
10H14O9N8P2
10H16O15N8P4
10
11
12
13
14
15
16
17a
17b
18
19
20
21
22
23
24
25
26
27
28
29a
29b
30
31
32
33
34
35
36
C9H14O11N2P2
10H16O11N2P2
C
C9H13O11N2P2F
C9H15O10N3P2
C
11H17O9N5P2
a
Purity of each derivative was 95%, as determined using HPLC (Hewlett-Packard 1090) with two different mobile phases, using an
SMT OD-5-60 RP-C18 column (250 × 4.6 mm; Separation Methods Technologies, Inc., Newark, DE). The flow rate was 1 mL/min, with
a typical pressure of 100 bar. System A: linear gradient solvent system, 0.1 M TEAA/CH3CN from 95/5 to 40/60 in 20 min. System B:
b
linear gradient solvent system, 5 mM TBAP/CH3CN from 80/20 to 40/60 in 20 min. The percent yields refer to phosphorylation reactions.
was 1 mL/min, with a typical pressure of 100 bar. System A:
linear gradient solvent system, 0.1 M TEAA/CH3CN from 95/5
to 40/60 in 20 min. System B: linear gradient solvent system,
5 mM TBAP/CH3CN from 80/20 to 40/60 in 20 min. Peaks were
detected by UV absorption using a diode array detector. All
derivatives showed more than 95% purity in HPLC system.
Low-resolution CI-NH3 (chemical ionization) mass spectra
were carried out with a Finnigan 4600 mass spectrometer and
high-resolution EI (electron impact) mass spectra with a
VG7070F mass spectrometry at 6 kV. High-resolution FAB
(fast atom bombardment) mass spectrometry was performed
(TEAA); elution time, 20 min; flow rate, 1 mL/min; column,
SMT OD-5-60 RP-C18; detector, UV, Emax ˜ 260-300 nm). The
reaction was quenched by adding 2 mL of triethylammonium
bicarbonate buffer and 3 mL of water. The mixture was
subsequently frozen and lyophilized. Purification was per-
formed on an ion-exchange column packed with Sephadex-
DEAE A-25 resin, linear gradient (0.01-0.5 M) of 0.5 M
ammonium bicarbonate was applied as the mobile phase, and
UV and HPLC were used to monitor the elution. All nucle-
otides were collected, frozen, and lyophilized as the ammonium
salts. All synthesized compounds gave correct molecular
masses and showed more than 95% purity (high-resolution
FAB, purity, and HPLC retention times are reported in Table
3).
with
a J EOL SX102 spectrometer using 6-kV Xe atoms
following desorbtion from a glycerol matrix.
Purification of most of the nucleotide analogues was carried
out on DEAE-A25 Sephadex columns as described below.
However, compounds 17a , 18, and 22 (system A) and com-
pounds 26 and 36 (system B followed by Sephadex column)
required HPLC purification (semipreparative C18 column) of
the reaction mixtures.
Gen er a l P r oced u r e of P h osp h or yla tion . Nucleoside (0.1
mmol) and Proton Sponge (107 mg, 0.5 mmol) were dried for
several hours in high vacuum at room temperature and then
suspended in 2 mL of trimethyl phosphate. Phosphorus
oxychloride (37 µL, 0.4 mmol) was added, and the mixture was
stirred for 1 h at 0 °C. The reaction was monitored by
analytical HPLC (eluting with gradient from buffer, CH3CN
) 95:5 to 40:60; buffer, 0.1 M triethylammonium acetate
Ad en in e Ar a b in ofu r a n osid e 3′,5′-Bis(d ia m m on iu m
p h osp h a te) (3). Starting from 40 mg (0.15 mmol) of adenine
arabinofuranoside and following the general procedure, using
an additional 2 mL of acetonitrile to dissolve starting material,
we obtained 11.2 mg (0.023 mmol, 15% yield) of 3: 1H NMR
(D2O) δ 4.09 (2H, m, CH2-5′), 4.33 (1H, bs, H-4′), 4.56 (1H, m,
H-2′), 4.65 (1H, m, H-3′), 6.43 (1H, d, J ) 3.9 Hz, H-1′), 8.18
(1H, s, H-2), 8.50 (1H, s, H-8); 31P NMR (D2O) δ 3.49 (s, 1-P),
3.70 (d, J ) 6.0 Hz, 1-P).
2′-Deoxy-1-d ea za -N6-m eth yla d en osin e 3′,5′-Bis(d ia m -
m on iu m p h osp h a te) (4). Starting from 14 mg (0.053 mmol)
of 2′-deoxy-1-deaza-N6-methyladenosine and following the
general procedure, we obtained 4.2 mg (0.009 mmol, 17% yield)

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1633
of 4: 1H NMR (D2O) δ 2.76 (2H, m, CH2-2′), 3.06 (3H, s,
NHCH3), 4.11 (2H, m, CH2-5′), 4.49 (1H, bs, H-4′), 4.97 (1H,
m, H-3′), 6.45 (1H, t, J ) 6.8 Hz, H-1′), 6.59 (1H, d, J ) 6.8
2.16-2.73 (1H, m, CH2-2′), 2.78-2.87 (1H, m, CH2-2′), 2.98
(3H, s, NHCH3), 3.07 (2H, t, J ) 6.8 Hz, CH3CH2CH2S), 4.03
(2H, m, CH2-5′), 4.40 (1H, bs, H-4′), 4.94 (1H, bs, H-3′) 6.40
(1H, t, J ) 5.8 Hz, H-1′), 8.20 (1H, s, H-8); 31P NMR (D2O) δ
0.86 (bs, 5′-P), 0.18 (bs, 3′-P).
Hz, H-1′), 8.02 (1H, d, J ) 6.8 Hz, H-2), 8.32 (1H, s, H-8); 31
NMR (D2O) δ 0.56 (d, J ) 16.6 Hz, 3′-P), 1.43 (bs, 5′-P).
P
2-Ch lor o-2′-d eoxy-N6-m eth yla d en osin e 3′,5′-Bis(d ia m -
m on iu m p h osp h a te) (6). Starting from 25 mg (0.083 mmol)
of 2-chloro-2′-deoxy-N6-methyladenosine, 39, and following the
general procedure, we obtained 7.5 mg (0.014 mmol, 17% yield)
of 6: 1H NMR (D2O) δ 2.76 (2H, m, CH2-2′), 2.95 (3H, s,
NHCH3), 4.06 (2H, m, CH2-5′), 4.43 (1H, bs, H-4′), 4.96 (1H,
m, H-3′), 6.32 (1H, t, J ) 6.9 Hz, H-1′), 8.31 (1H, s, H-8); 31P
NMR (D2O) δ 0.06 (d, J ) 9.0 Hz, 3′-P), 0.73 (bs, 5′-P).
2′-Deoxy-2-iod o-N6-m eth yla d en osin e 3′,5′-Bis(d ia m m o-
n iu m p h osp h a te) (7). Starting from 28 mg (0.072 mmol) of
2′-deoxy-2-iodo-N6-methyladenosine, 50, and following the
general procedure, we obtained 6.6 mg (0.012 mmol, 17% yield)
of 7: 1H NMR (D2O) δ 2.31 (1H, m, CH2-2′), 2.97 (1H, m, CH2-
2′), 3.13 (3H, s, NHCH3), 4.11 (2H, m, CH2-5′), 4.46 (1H, m,
H-4′), 5.06 (1H, m, H-3′), 6.65 (1H, t, J ) 5.9 Hz, H-1′), 8.65
(1H, s, H-8); 31P NMR (D2O) δ 0.56 (bs, 5′P), -0.19 (bs, 3′P).
2′-Deoxy-N6-m eth oxya d en osin e 3′,5′-Bis(d ia m m on iu m
p h osp h a te) (8). Starting from 12 mg (0.043 mmol) of 2′-deoxy-
N6-methoxyladenosine, 59, and following the general proce-
dure, we obtained 5.1 mg (0.010 mmol, 23% yield) of 8: 1H
NMR (D2O) δ 2.75 (2H, m, CH2-2′), 3.84 (3H, s, OCH3), 4.04
(2H, m, CH2-5′), 4.43 (1H, bs, H-4′), 4.97 (1H, m, H-3′), 6.45
(1H, t, J ) 6.9 Hz, H-1′), 7.95 (1H, s, H-2), 8.30 (1H, s, H-8);
31P NMR (D2O) δ 0.60 (d, J ) 7.5 Hz, 3′-P), 1.13 (bs, 5′-P).
2′-Deoxy-2-(m eth ylth io)-N6-m eth yla d en osin e 3′,5′-Bis-
(d ia m m on iu m p h osp h a te) (10). Starting from 25 mg (0.083
mmol) of 2′-deoxy-2-(methylthio)-N6-methyladenosine, 40, and
following the general procedure, we obtained 7.5 mg (0.014
mmol, 17% yield) of 10: 1H NMR (D2O) δ 2.55 (3H, s, SCH3),
2.67-286 (2H, m, CH2-2′), 3.02 (3H, s, NHCH3), 3.95 (2H, m,
CH2-5′), 4.37 (1H, bs, H-4′), 4.88 (1H, m, H-3′), 6.47 (1H, t, J
) 6.9 Hz, H-1′), 8.29 (1H, s, H-8); 31P NMR (D2O) δ 0.06 (d, J
) 9.0 Hz, 3′-P), 0.73 (bs, 5′-P).
2′-Deoxy-2-(eth ylth io)a d en osin e 3′,5′-Bis(d ia m m on iu m
p h osp h a te) (11). Starting from 22 mg (0.071 mmol) of
2′-deoxy-2-(ethylthio)adenosine, 52, and following the general
procedure, we obtained 7.6 mg (0.016 mmol, 23% yield) of 11:
1H NMR (D2O) δ 1.50 (3H, t, J ) 7.3 Hz, CH3CH2S), 2.85 (1H,
m, CH2-2′), 3.05 (1H, m, CH2-2′), 3.32 (2H, q, J ) 7.81 Hz,
CH3CH2S), 4.18 (2H, m, CH2-5′), 4.58 (1H, m, H-4′), 5.10 (1H,
m, H-3′), 6.72 (1H, t, J ) 6.5 Hz, H-1′), 8.49 (1H, s, H-8); 31P
NMR (D2O) δ 1.48 (d, J ) 9.1 Hz, 5′-P), 1.02 (d, J ) 14.1 Hz,
3′-P).
2′-Deoxy-2-(eth ylth io)-N6-m eth yla d en osin e 3′,5′-Bis(d i-
a m m on iu m p h osp h a te) (12). Starting from 20 mg (0.061
mmol) of 2′-deoxy-2-(ethylthio)-N6-methyladenosine, 41, and
following the general procedure, we obtained 6.8 mg (0.014
mmol, 23% yield) of 12: 1H NMR (D2O) δ 1.49 (3H, t, J ) 7.7
Hz, CH3CH2S), 2.85-3.01 (2H, m, CH2-2′), 3.22 (3H, s,
NHCH3), 3.32 (2H, q, J ) 7.6 Hz, CH3CH2S), 4.06 (2H, m, CH2-
5′), 4.56 (1H, m, H-4′), 5.12 (1H, m, H-3′), 6.69 (1H, t, J ) 6.8
Hz, H-1′), 8.55 (1H, s, H-8); 31P NMR (D2O) δ 3.97 (s, 5′-P),
3.43 (s, 3′-P).
2-(Allylth io)-2′-d eoxy-N6-m eth yla d en osin e 3′,5′-Bis(d i-
a m m on iu m p h osp h a te) (15). Starting from 20 mg (0.084
mmol) of 2-(allylthio)-2′-deoxy-N6-methyladenosine, 43, and
following the general procedure, we obtained 5.8 mg (0.117
mmol, 13% yield) of 15: 1H NMR (D2O) δ 2.89 (2H, m, CH2-
2′), 3.17 (3H, s, NHCH3), 3.97 (2H, m, CH2dCHCH2S), 4.17
(2H, m, CH2-5′), 4.55 (1H, m, H-4′), 5.09 (1H, m, H-3′), 5.27
(1H, d, J ) 10.0 Hz, CH2dCHCH2S), 5.51 (1H, d, J ) 12.0 Hz,
CH2dCHCH2S), 6.20 (1H, m, CH2dCHCH2S), 6.60 (1H, t, J )
6.0 Hz, H-1′), 8.39 (1H, s, H-8); 31P NMR (D2O) δ 1.31 (bs, 5′-
P), 0.78 (bs, 3′-P).
2′-Deoxy-2-(h exylt h io)-N6-m et h yla d en osin e 3′,5′-Bis-
(d ia m m on iu m p h osp h a te) (16). Starting from 28 mg (0.073
mmol) of 2′-deoxy-2-(hexylthio)-N6-methyladenosine, 44, and
following the general procedure, we obtained 10.8 mg (0.018
mmol, 25% yield) of 16: 1H NMR (D2O) δ 0.82 (3H, m, CH3-
CH2), 1.26 (4H, m, CH2CH2), 1.40 (2H, m, CH2), 1.72 (2H, m,
CH2CH2S), 2.68-2.88 (2H, m, CH2-2′), 3.05 (3H, bs, NHCH3),
3.15 (2H, m, CH2S), 4.03 (2H, m, CH2-5′), 4.40 (1H, bs, H-4′),
4.94 (1H, m, H-3′), 6.47 (1H, t, J ) 6.8 Hz, H-1′), 8.30 (1H, s,
H-8); 31P NMR (D2O) δ 1.44 (s, 3′-P); 1.90, (s, 5′-P).
2-Am in o-2′-d eoxy-N6-m eth yla d en osin e 3′,5′-Bis(d ia m -
m on iu m p h osp h a te) (17a ). Starting from 15 mg (0.053
mmol) of 2-amino-2′-deoxy-N6-methyladenosine, 48, and fol-
lowing the general procedure, we obtained 4.4 mg (0.01 mmol,
19% yield) of 17a : 1H NMR (D2O) δ 2.50 (1H, m, CH2-2′), 2.67
(1H, m, CH2-2′), 2.99 (1H, s, NHCH3), 3.96 (2H, m, CH2-5′),
4.21 (1H, m, H-4′), 4.85 (1H, m, H-3′), 6.27 (1H, t, J ) 6.6 Hz,
H-1′), 8.06 (1H, s, H-8); 31P NMR (D2O) δ 0.99 (s, 5′-P), 0.49
(bs, 3′-P).
2′-Deoxygu a n osin e 3′,5′-Bis(d ia m m on iu m p h osp h a te)
(17b). Starting from 50 mg (0.22 mmol) of 2′-deoxyguanosine
and following the general procedure, we obtained 39.4 mg
(0.086 mmol, 39% yield) of 17b: 1H NMR (D2O) δ 2.40 (1H,
m, CH2-2′), 2.55 (1H, m, CH2-2′), 4.08 (2H, m, CH2-5′), 4.36
(1H, bs, H-4′), 4.80 (H-3′ under D2O peak), 5.93 (1H, d, J )
8.8 Hz, H-5), 6.37 (1H, t, J ) 6.8 Hz, H-1′), 7.98 (1H, d, J )
8.8 Hz, H-6); 31P NMR (D2O) δ 0.89, (s, 5′-P); 0.31 (s, 3′-P).
2′-Deoxy-8-m et h yla d en osin e 3′,5′-Bis(d ia m m on iu m
p h osp h a te) (18). Starting from 12.0 mg (0.045 mmol) of
8-methyl-2′-deoxyadenosine30 and following the general pro-
cedure, we obtained 2.0 mg (0.004 mmol, 8.9% yield) of 18:
1H NMR (D2O) δ 2.50-2.59 (1H, m, CH2-2′), 3.02-3.20 (1H,
m, CH2-2′), 2.68 (3H, s, 8-CH3), 4.06-4.10 (2H, m, CH2-5′), 4.30
(1H, bs, H-4′), 4.90 (1H, m, H-3′), 6.49 (1H, t, J ) 6.8 Hz, H-1′),
8.20 (1H, s, H-2); 31P NMR (D2O) δ 1.88 (bs, 5′-P), 1.62 (bs,
3′-P).
2′-Deoxy-8-vin yladen osin e 3′,5′-Bis(diam m on iu m ph os-
p h a te) (19). Starting from 15.0 mg (0.054 mmol) of 8-vinyl-
2′-deoxyadenosine30 and following the general procedure, we
obtained 3.7 mg (0.0073 mmol, 13.6% yield) of 19: 1H NMR-
(D2O) δ 2.50-2.55 (1H, m, CH2-2′), 2.95-3.10 (1H, m, CH2-
2′), 4.01-4.05 (2H, m, CH2-5′), 4.30 (1H, bs, H-4′), 4.91 (1H,
m, H-3′), 5.84 (1H, d, J ) 10.7 Hz, vinyl), 6.37 (1H, d, J )
16.6 Hz, vinyl), 6.56 (1H, t, J ) 5.8 Hz, H-1′), 7.20 (1H, dd, J
) 16.6, 10.7 Hz, vinyl), 8.20 (1H, s, H-2); 31P NMR (D2O) δ
3.85 (bs, 5′-P), 3.36 (bs, 3′-P).
2′-Deoxy-8-O-m eth yla d en osin e 3′,5′-Bis(d ia m m on iu m
p h osp h a te) (20). Starting from 24 mg (0.085 mmol) of
2′-deoxy-8-O-methyl-adenosine, 56, and following the general
procedure, we obtained 4.2 mg (0.009 mmol, 17% yield) of 20:
1H NMR (D2O) δ 2.42 (1H, m, CH2-2′), 3.20 (1H, m, CH2-2′),
3.89 (3H, s, OCH3), 4.0-4.35 (3H, m, CH2-5′ and H-4′), 5.1 (1H,
m, H-3′), 6.61 (1H, t, J ) 6.4 Hz, H-1′), 8.10 (1H, s, H-2); 31P
NMR (D2O) δ 0.78 (s, 5′-P), 0.25 (bs, 3′-P).
2′-Deoxy-2-(p r op ylth io)a d en osin e 3′,5′-Bis(d ia m m on i-
u m p h osp h a te) (13). Starting from 39.6 mg (0.122 mmol) of
2-(propylthio)-2′-deoxyadenosine, 53, and following the general
procedure, we obtained 14.2 mg (0.026 mmol, 21.3% yield) of
13: 1H NMR(D2O) δ 0.98 (3H, t, J ) 6.8 Hz, CH3CH2CH2S),
1.66-1.78 (2H, m, CH3CH2CH2S) 2.69-2.75 (1H, m, CH2-2′),
2.81-2.93 (1H, m, CH2-2′), 3.16 (2H, t, J ) 6.8 Hz, CH3-
CH2CH2S), 4.06 (2H, m, CH2-5′), 4.43 (1H, bs, H-4′), 4.97(1H,
bs, H-3′), 6.51 (1H, t, J ) 6.8 Hz, H-1′), 8.32 (1H, s, H-8); 31P
NMR (D2O) 0.54 (s, 5′-P), -0.22 (d, J ) 1.8 Hz, 3′-P).
2′-Deoxy-2-(p r op ylth io)-N6-m eth yla d en osin e 3′,5′-Bis-
(d ia m m on iu m p h osp h a te) (14). Starting from 31.0 mg
(0.091 mmol) of N6-methyl-2-(propylthio)-2′-deoxyadenosine,
42, and following the general procedure, we obtained 5.0 mg
(0.009 mmol, 10% yield) of 14: 1H NMR (D2O) δ 0.96 (3H, t, J
) 6.8 Hz, CH3CH2CH2S), 1.64-1.71 (2H, m, CH3CH2CH2S)
2′-Deoxy-8-(m eth ylth io)a d en osin e 3′,5′-Bis(d ia m m on i-
u m p h osp h a te) (21). Starting from 25 mg (0.084 mmol) of
2′-deoxy-8-(methylthio)adenosine, 57, and following the gen-
eral procedure, we obtained 14.2 mg (0.031 mmol, 37% yield)

1634 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
of 21: 1H NMR (D2O) δ 2.45 (1H, m, CH2-2′), 2.72 (3H, s,
SCH3), 3.21 (1H, m, CH2-2′), 4.08 (2H, m, CH2-5′), 4.25 (1H,
m, H-4′), 5.0 (1H, m, H-3′), 6.5 (1H, t, J ) 6.5 Hz, H-1′), 8.15
(1H, s, H-2); 31P NMR (D2O) δ 0.98 (bs, 5′-P), 0.31 (bs, 3′-P).
2′-Deoxy-8-(m eth yla m in o)a d en osin e 3′,5′-Bis(d ia m m o-
n iu m p h osp h a te) (22). Starting from 28.1 mg (0.1 mmol) of
2′-deoxy-8-(methylamino)adenosine, 58, and following the
general procedure, we obtained 2.4 mg (0.0055 mmol, 5.5%
yield) of 22: 1H NMR (D2O) δ 2.53 (1H, m, CH2-2′), 3.09 (3H,
s, NHCH3), 3.24 (1H, m, CH2-2′), 4.12 (2H, m, CH2-5′), 4.30
(1H, m, H-4′), 4.99 (1H, m, H-3′), 6.47 (1H, t, J ) 7.4 Hz, H-1′),
8.15 (1H, s, H-2); 31P NMR (D2O) δ 0.91 (s, 5′-P), 0.25 (bs, 3′-
P).
2′-Deoxy-4′-th ioa d en osin e 3′,5′-Bis(d ia m m on iu m p h os-
p h a te) (23). Starting from 12 mg (0.045 mmol) of 2′-deoxy-
4′-thioadenosine, 62, and following the general procedure, we
obtained 2.4 mg (0.0056 mmol, 13% yield) of 23: 1H NMR
(D2O) δ 2.59 (2H, m, CH2-2′), 3.97 (2H, m, CH2-5′), 4.32 (1H,
m, H-4′), 4.95 (1H, m, H-3′), 6.30 (1H, t, J ) 7.0 Hz, H-1′),
8.06 (1H, s, H-2), 8.18 (1H, s, H-8); 31P NMR (D2O) δ 0.55 (s,
5′-P), -0.20 (d, J ) 2.4 Hz, 3′-P).
(1R,2R,4R)-4-(9H-Ad en in -9-yl)-2-(p h osp h a tom eth yl)cy-
clop en ta n -1-yl P h osp h a te (24). Starting from 6.0 mg (0.024
mmol) of carbocyclic 2′-deoxyadenosine, 56,27 and following the
general procedure, we obtained 2.4 mg (0.0059 mmol, 24.5%
yield) of 24, the carbocyclic equivalent of 2′-deoxyadenosine
3′,5′-bis(diammonium phosphate): 1H NMR (D2O) δ 1.98 (1H,
m, CH2-2′), 2.5 (3H, m, CH2-2′ and CH2-6′), 4.03 (2H, m, CH2-
5′), 4.68 (1H, m, H-4′), 5.12 (1H, m, H-3′), 5.25 (1H, m, J ) 5.5
Hz, H-1′), 8.28 (1H, s, H-2), 8.34 (1H, s, H-8); 31P NMR (D2O)
δ 0.94 (s, 5′-P), -0.07 (s, 3′-P).
(29b). A solution of 20 mg (0.04 mmol) of 2 in anhydrous DMF
was evaporated under reduced pressure below 30 °C, and this
drying procedure was repeated three times. The residue was
resuspended in 2 mL of anhydrous DMF under nitrogen
atmosphere, and 38 µL (0.16 mmol) of tributylamine was added
at room temperature. The mixture was stirred for 30 min, and
a solution of 33 mg (0.2 mmol) of carbonyldiimidazole was
added. The mixture was stirred for 12 h at room temperature
before the excess carbonyldiimidazole was quenched by addi-
tion of 0.05 mL of methanol, and the mixture was stirred for
30 min. A solution of 0.091 g of bis(tributylammonium)-
pyrophosphate (0.2 mmol) in 2 mL of anhydrous DMF was
added at room temperature, and the mixture was stirred for
42 h. The reaction mixture was evaporated under reduced
pressure below 30 °C, and two products were purified by the
ion-exchange column chromatography in the general procedure
and semipreparative HPLC using SMT OD-5-60 C18 column
with a linear gradient elution of 1.0 M triethylammonium
acetate solution and CH3CN (95/5-40/60 for 30 min). After
evaporation of the pure fraction under reduced pressure, the
products were redissolved in 0.5 M ammonium bicarbonate
solution, and repeated lyophilization with water gave 3.5 mg
of 29a (yield 20%) and 1.2 mg of 29b (yield 5%).
29a : 1H NMR (D2O) δ 2.74-2.84 (1H, m, CH2-2′), 2.97-3.02
(1H, m, CH2-2′), 3.04 (3H, s, NHCH3), 3.83-3.96 (1H, m, H-4′),
4.21-4.39 (2H, m, H-5′), 5.19 (1H, pen, J ) 7.8 Hz, H-3′), 6.39-
6.42 (1H, m, H-1′), 8.20 (1H, s, H-2), 8.23 (1H, s, H-8); 31P NMR
(D2O) δ -12.45 (dd, J ) 8.5, 24.5 Hz, 3′-CHOP-), -11.71 (2,
dd, J ) 10.3, 18.6, 24.4 Hz, 5′-CH2OP-), with proton decoupling
off mode.
29b: 1H NMR (D2O) δ 2.75-2.84 (1H, m, CH2-2′), 2.90-2.99
(1H, m, CH2-2′), 3.09 (3H, s, NHCH3), 4.10-4.16 (1H, m, H-4′),
4.29-4.40 (2H, m, H-5′), 5.27 (1H, pen, J ) 6.8 Hz, H-3′), 6.50
(1H, t, J ) 5.8 Hz, H-1′), 8.26 (1H, s, H-2), 8.42 (1H, s, H-8);
31P NMR (D2O) δ -23.38 to -23.22 (2P, m), -12.01 to -11.80
(m), -11.05 to -10.95 (m), with proton decoupling off mode;
Negative API-ES Mass 566 (M - H).
3′-Azid o-3′-d eoxyxyloa d en osin e 2′,5′-Bis(d ia m m on iu m
p h osp h a te) (30). Starting from 25 mg (0.085 mmol) of
3′-azido-3′-deoxyxyloadenosine and following the general pro-
cedure, we obtained 4.9 mg (0.011 mmol, 13% yield) of 30: 1H
NMR (D2O) δ 3.1 (1H, m, H-2′), 3.64 (1H, m, H-3′), 4.30 (2H,
m, CH2-5′), 4.53 (1H, m, H-4′), 6.12 (1H, bs, H-1′), 8.25 (1H, s,
H-2), 8.48 (1H, s, H-8); 31P NMR (D2O) δ 1.37 (bs, 5′-P), 0.55
(bs, 3′-P).
3′-Azid o-3′-d eoxyxyloa d en osin e 2′-Mon op h osp h a te 5′-
Tr ip h osp h a te (h exa a m m on iu m sa lt) (31). Starting from
25 mg (0.085 mmol) of 3′-azido-3′-deoxyxyloadenosine and
following the general procedure, we obtained 1.7 mg (0.0028
mmol, 3.0% yield) of 31: 1H NMR (D2O) δ 3.14 (1H, m, H-2′),
3.65 (1H, m, H-3′), 4.35 (2H, m, CH2-5′), 4.56 (1H, M, H-4′),
6.13 (1H, bs, H-1′), 8.28 (1H, s, H-2), 8.51 (1H, s, H-8); 31P
NMR (D2O) δ 0.43 (s, 3′-P), -10.43 (d, J ) 19.5 Hz, 5′-Pγ),
-11.11, -11.27 (2t, J ) 6.1 Hz, 5′-Pâ), -22.76 (t, J ) 19.7 Hz,
5′-PR).
2′-Deoxyu r idin e 3′,5′-Bis(diam m on iu m ph osph ate) (32).
Starting from 50 mg (0.22 mmol) of 2′-deoxyuridine and
following the general procedure, we obtained 39.4 mg (0.086
mmol, 39% yield) of 32: 1H NMR (D2O) δ 2.40 (1H, m, CH2-
2′), 2.55 (1H, m, CH2-2′), 4.08 (2H, m, CH2-5′), 4.36 (1H, bs,
H-4′), 4.80 (H3′ under D2O peak), 5.93 (1H, d, J ) 8.8 Hz, H-5),
6.37 (1H, t, J ) 6.8 Hz, H-1′), 7.98 (1H, d, J ) 8.8 Hz, H-6);
31P NMR (D2O) δ 0.89 (s, 5′-P), 0.31 (s, 3′-P).
Th ym idin e 3′,5′-Bis(diam m on iu m ph osph ate) (33). Start-
ing from 50 mg (0.22 mmol) of thymidine and following the
general procedure, we obtained 39.4 mg (0.086 mmol, 39%
yield) of 33: 1H NMR (D2O) δ 2.40 (1H, m, CH2-2′), 2.55 (1H,
m, CH2-2′), 4.08 (2H, m, CH2-5′), 4.36 (1H, bs, H4′), 4.80 (H3′
under D2O peak), 5.93 (1H, d, J ) 8.8 Hz, H5), 6.37 (1H, t, J
) 6.8 Hz, H1′), 7.98 (1H, d, J ) 8.8 Hz, H6); 31P NMR (D2O)
δ 1.04 (s, 5′-P), 0.75 (s, 3′-P).
(1R ,2R ,4R )-4-(9H -2-Am in o-N 6-m e t h yla d e n in -9-yl)-2-
(ph osph atom eth yl)cyclopen tan -1-yl P h osph ate (25). Start-
ing from 15 mg (0.054 mmol) of the carbocyclic equivalent of
2-amino-2′-deoxy-N6-methyladenosine, 61, and following the
general procedure, we obtained 2.4 mg (0.0055 mmol, 10%
yield) of 25, the carbocyclic equivalent of 2-amino-2′-deoxy-
N6-methyladenosine 3′,5′-bis(diammonium phosphate): 1H
NMR (D2O) δ 1.87 (1H, m, CH2-2′), 2.29 (2H, m, CH2-6′), 2.54
(1H, m, CH2-2′), 3.13 (3H, s, NHCH3), 3.99 (2H, m, CH2-5′),
4.40 (1H, m, H-4′), 4.96 (2H, m, H-3′ and H-1′), 8.00 (1H, s,
H-8); 31P NMR (D2O) δ 0.92 (bs, 5′-P), 0.42 (s, 3′-P).
2′-Deoxy-4′-th ioa d en osin e S-Oxid e 5′-Mon o(d ia m m o-
n iu m p h osp h a te) (26). Starting from 14.0 mg (0.049 mmol)
of 2′-deoxy-4′-thioadenosine S-oxide, 63, and following the
general procedure, we obtained 0.50 mg (0.001 mmol, 2.0%
yield) of 26: 1H NMR (D2O) δ 2.79-2.95 (1H, m, H-2′), 3.11-
3.27 (1H, m, H-2′), 3.51-3.66 (2H, m, H-5′), 3.72-3.77 (1H,
m, H-4′), 4.29-4.34 (1H, m, H- 3′) 6.18-6.38 (1H, m, H-1′),
8.30 (1H, s, H-2) 8.37 (1H, s, H-8); 31P NMR (D2O) δ 0.46, 0.34
(2s, mixture of diastereomers).
2′-Deoxy-5′-O-p h osp h on ylm et h yla d en osin e 3′-P h os-
p h a te (tetr a m m on iu m sa lt) (27). Starting from 15 mg (0.043
mmol) of 2′-deoxy-5′-[diethyl(methylphosphonyl)]adenosine, 64,
and following the general procedure, we obtained 1.8 mg
(0.0042 mmol, 9.8% yield) of 27. Cleavage of the phosphonate
esters was verified by NMR and MS: 1H NMR (D2O) δ 2.83
(2H, m, CH2-2′), 4.10 (4H, m, CH2-5′ and CH2-P), 4.47 (1H,
m, H-4′), 5.10 (1H, m, H-3′), 6.58 (1H, t, J ) 7.9 Hz, H-1′),
8.28 (1H, s, H-2), 8.51 (1H, s, H-8); 31P NMR (D2O) δ 27.49 (s,
5′-CH2P), 3.31 (s, 3′-P).
2′-Deoxya d en osin e 3′,5′-Bis(a m m on iu m m eth yl p h os-
p h a te) (28). Starting from 25 mg (0.1 mmol) of 2′-deoxyad-
enosine and following the general procedure using methyl
dichlorophosphate (59.6 mg, 41.13 µL, 0.4 mmol), we obtained
9.8 mg (0.018 mmol, 22% yield) of 28: 1H NMR (D2O) δ 2.64
(1H, m, CH2-2′), 2.86 (1H, m, CH2-2′), 3.42 (3H, d, J ) 10.8
Hz, CH3OP), 3.58 (3H, d, J ) 10.8 Hz, CH3OP), 4.02 (2H, m,
CH2-5′), 4.28 (1H, m, H-4′), 4.53 (1H, m, H-3′), 6.46 (1H, t, J
) 6.9 Hz, H-1′), 8.18 (1H, s, H-2), 8.37 (1H, s, H-8); 31P NMR
(D2O) δ 3.30 (bs, 5′-P), 1.79 (d, J ) 2.4 Hz, 3′-P).
2′-Deoxy-N6-m eth yla d en osin e 3′,5′-Cyclic-d ip h osp h a te
(d ia m m on iu m sa lt) (29a ) a n d 2′-Deoxy-N6-m eth yla d -
en osin e 3′,5′-Cyclic-tetr aph osph ate (tetr aam m on iu m salt)
5-F lu or o-2′-d eoxyu r id in e 3′,5′-Bis(d ia m m on iu m p h os-
p h a te) (34). Starting from 20 mg (0.081 mmol) of 5-fluoro-2′-
deoxyuridine and following the general procedure, we obtained

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1635
8.9 mg (0.022 mmol, 27% yield) of 34: 1H NMR (D2O) δ 2.38
(1H, m, CH2-2′), 2.55 (1H, m, CH2-2′), 4.08 (2H, m, CH2-5′),
4.35 (1H, m, H-4′), 4.94 (1H, m, H-3′), 6.35 (1H, bs, H-1′), 8.17
(1H, s, H-5); 31P NMR (D2O) δ 1.52 (s, 5′-P), 1.19 (s, 3′-P).
2′-Deoxycytidin e 3,5′-Bis(diam m on iu m ph osph ate) (35).
Starting from 20 mg (0.087 mmol) of 2′-deoxycytidine and
following the general procedure, we obtained 12 mg (0.031
mmol, 36% yield) of 35: 1H NMR (D2O) δ 2.41 (1H, m, CH2-
2′), 2.65 (1H, m, CH2-2′), 4.11 (2H, m, CH2-5′), 4.42 (1H, bs,
H-4′), 4.85 (1H, m, H-3′), 6.37 (1H, t, J ) 6.2 Hz, H-1′), 8.07
(1H, d, J ) 8.0 Hz, 1H), 8.18 (d, J ) 7.5 Hz, 1H); 31P NMR
(D2O) δ 0.51 (s, 5′-P), -0.13 (s, 3′-P).
1′,5′-An h yd r o-2′-(a d en in -9-yl)-2′,3′-d id eoxy-D-a r a bin o-
h exitol 4′,6′-Bis(d ia m m on iu m p h osp h a te) (36). Starting
from 19.2 mg (0.072 mmol) of 1′,5′-anhydro-2′-(adenin-9-yl)-
2′,3′-dideoxy-D-arabino-hexitol25 and following the general
phosphorylation procedure, we obtained 0.50 mg (0.001 mmol,
1.4% yield) of 36: 1H NMR (D2O) δ 2.04-2.74 (2H, m, CH2-
3′), 3.57-3.60 (1H, m, H-5′), 3.69-3.72 (1H, m, H-4′), 4.09 (2H,
bs, CH2-6′), 4.11-4.33 (2H, m, CH2-1′), 4.93 (1H, bs, H-2′), 8.23
(1H, s, H-2), 8.51 (1H, s, H-8); 31P NMR (D2O) δ 3.28, 1.75
(2s).
CH2-2′), 3.10 (3H, s, N-CH3), 3.22 (2H, q, J ) 6.4 Hz,
CH3CH2S), 3.77-3.90 (2H, m, CH2-5′), 4.09 (1H, bs, H-4′), 4.64
(1H, bs, H-3′), 6.44 (1H, t, J ) 5.9 Hz, H-1′), 8.03 (2H, bs, NH2),
8.23 (1H, s, H-8).
2-(P r op ylth io)-N6-m eth yl-2′-d eoxya d en osin e (42). To a
suspension of potassium tert-butoxide (112 mg, 1 mmol) in dry
DMF (3 mL) in a sealed tube was added propanethiol (72 mg,
86 µL, 1 mmol), and the reaction mixture was stirred at room
temperature for 30 min. 2-Chloro-2′-deoxy-N6-methyladenosine
(39; 50 mg, 0.17 mmol) in dry DMF (1 mL) was added and
heated at 110 °C for 4 h while stirring. The solvent was
removed under reduced pressure, and the residue was chro-
matographed using pTLC (CHCl3/MeOH, 9/1) to provide 42
as a colorless solid (46 mg, 0.14 mmol, 80% yield): MS (CI-
1
NH3) 340 (M+ + 1); H NMR (CD3OD) δ 1.13 (3H, t, J ) 6.8
Hz, CH3CH2CH2S), 1.83-1.87 (2H, m, CH3CH2CH2S), 2.39-
2.42 (1H, m, CH2-2′), 2.84-2.92 (1H, m, CH2-2′), 3.17 (3H, s,
N-CH3), 3.19 (2H, bs, CH3CH2CH2S), 3.80-3.92 (2H, m, CH2-
5′), 4.12 (1H, bs, H-4′), 4.62 (1H, bs, H-3′), 6.42 (1H, t, J ) 6.4
Hz, H-1′), 8.17 (1H, s, H-8).
2-(Allylth io)-N6-m eth yl-2′-d eoxya d en osin e (43). To a
suspension of potassium tert-butoxide (112 mg, 1 mmol) in dry
DMF (3 mL) in a sealed tube was added allylthiol (70 mg, 1
mmol), and the reaction mixture was stirred at room temper-
ature for 30 min. 2-Chloro-2′-deoxy-N6-methyladenosine (39;
50 mg, 0.17 mmol) in dry DMF (1 mL) was added and heated
at 110 °C for 4 h while stirring. The solvent was removed under
reduced pressure, and the residue was chromatographed using
pTLC (CHCl3/MeOH, 9/1) to provide 43 as a colorless solid (41
mg, 0.12 mmol, 72% yield): MS (CI-NH3) 338 (M+ + 1); 1H
NMR (CD3OD) δ 2.32-2.37 (1H, m, CH2-2′), 2.72-2.80 (1H,
m, CH2-2′), 3.09 (3H, s, NCH3), 3.68-3.72 (2H, m, CH2d
CHCH2S), 4.01 (1H, bs, H-4′), 4.55 (1H, m, H-3′), 5.05 (1H, d,
J ) 10.5 Hz, CH2dCHCH2S), 5.31 (1H, d, J ) 12.6 Hz, CH2d
CHCH2S), 5.80 (1H, m, CH2)CHCH2S), 6.35 (1H, t, J ) 5.6
Hz, H-1′), 8.11 (1H, s, H-8).
2-(Hexylth io)-N6-m eth yl-2′-d eoxya d en osin e (44). To a
suspension of potassium tert-butoxide (146 mg, 1.3 mmol) in
dry DMF (3 mL) in a sealed tube was added hexanethiol (183
µL, 1.3 mmol), and the reaction mixture was stirred at room
temperature for 30 min. To the reaction mixture was added a
solution of 2-chloro-2′-deoxy-N6-methyladenosine (39; 38 mg,
0.13 mmol) in dry DMF (2 mL), and the mixture was stirred
and heated at 110 °C for 4 h. The solvent was removed under
reduced pressure, and the residue was chromatographed using
pTLC (CHCl3/MeOH, 9/1) to obtain the desired compound 2′-
deoxy-2-(hexylthio)-N6-methyladenosine (44) as a colorless
solid (35 mg, 0.091 mmol, 70% yield): MS (CI-NH3) 382 (M+
+ 1); 1H NMR (CD3OD) δ 0.87 (3H, m, CH3CH2), 1.29 (4H, m,
CH2CH2), 1.38 (2H, m, CH2), 1.67 (2H, m, CH2CH2S), 2.29-
2.37 (1H, m, CH2-2′), 2.68-2.78 (1H, m, CH2-2′), 3.05 (3H, bs,
NHCH3), 3.10 (2H, m, CH2S), 3.76 (2H, m, CH2-5′), 4.00 (1H,
m, H-4′), 4.51 (1H, m, H-3′), 6.31 (1H, t, J ) 6.8 Hz, H-1′),
8.05 (1H, s, H-8).
2-Ch lor o-2′-deoxy-3′,5′-di-O-acetyladen osin e (38). A mix-
ture of commercial 2-chloro-2′-deoxyadenosine (37; 285 mg, 1
mmol), pyridine (3 mL), and acetic anhydride (2 mL) was
stirred at room temperature for 2 h under nitrogen atmo-
sphere. The reaction mixture was evaporated to dryness and
chromatographed by pTLC (CHCl3/MeOH, 95/5) to obtain 38
as a white solid (300 mg, 0.81 mmol, 81% yield): MS (CI-NH3)
370 (M+ + 1), 387 (M + NH4+); H NMR (CDCl3) δ 2.11 (3H,
1
s, CH3CO), 2.14 (3H, s, CH3CO), 2.61-2.69 (1H, m, CH2-2′),
2.76-2.86 (1H, m, CH2-2′), 4.38 (3H, m, CH2-5′ and H-4′), 5.39
(1H, m, H-3′), 6.04 (2H, s, NH2), 6.45 (1H, t, J ) 6.8 Hz, H-1′),
7.99 (1H, s, H-8).
2-Ch lor o-N6-m eth yl-2′-d eoxya d en osin e (39). A flame
dried 25-mL round-bottom flask fitted with a reflux condenser
was charged with 38 (250 mg, 0.68 mmol), isoamyl nitrite (2
mL), and diiodomethane (2 mL) under nitrogen atmosphere.
The mixture was stirred at 85 °C for 1 h. The solvent was
removed with nitrogen purging, and the crude residue was
stirred with aqueous methylamine (5 mL, 40% solution) for 1
h. The water was removed under reduced pressure, and the
residue was chromatographed using pTLC (CHCl3/MeOH, 9/1)
to furnish 39 as a colorless solid (85.4 mg, 0.29 mmol, 42%
yield): MS (CI-NH3) 300 (M+ + 1); 1H NMR (CD3OD) δ 2.35-
2.43 (1H, m, CH2-2′), 2.68, 2.68-2.80 (1H, m, CH2-2′), 3.06 (3H,
bs, NHCH3), 3.10 (2H, m, CH2S), 3.79 (2H, m, CH2-5′), 4.04
(1H, m, H-4′), 4.57 (1H, m, H-3′), 6.34 (1H, t, J ) 6.8 Hz, H-1′),
8.20 (1H, s, H-8).
2-(Meth ylth io)-N6-m eth yl-2′-d eoxya d en osin e (40). To a
solution of 2-chloro-2′-deoxy-N6-methyladenosine (39; 50 mg,
0.17 mmol) in dry DMF (3 mL) was added sodium thiomethox-
ide (35 mg, 0.5 mmol), and the reaction mixture was stirred
and heated in a sealed tube at 110 °C for 4 h. The solvent was
removed under reduced pressure, and the residue was chro-
matographed using pTLC (CHCl3/MeOH, 9/1) to give 40 as a
colorless solid (40 mg, 0.13 mmol, 76% yield): MS (CI-NH3)
312 (M+ + 1); 1H NMR (CD3OD) δ 2.35-2.42 (1H, m, CH2-2′),
2.65 (3H, s, CH3S), 2.78-2.88 (1H, m, CH2-2′), 3.09 (3H, bs,
NHCH3), 3.71-3.84 (2H, m, CH2-5′), 4.01-4.03 (1H, m, H-4′),
4.56-4.58 (1H, m, H-3′), 6.36 (1H, t, J ) 6.8 Hz, H-1′), 8.09
(1H, s, H-8).
2′-Deoxy-3′,5′-d i-O-a cetylgu a n osin e (46). A mixture of
commercial 2′-deoxyguanosine (267 mg, 1 mmol), pyridine (4
mL), and acetic anhydride (2 mL) was heated while stirring
at 65 °C for 4 h under nitrogen atmosphere. The reaction
mixture was cooled, quenched with methanol, and evaporated
to dryness. The crude residue was purified by flash column
chromatography (CHCl3/MeOH, 95/5) to obtain 46 as a white
solid (263 mg, 0.72 mmol, 75% yield): MS (CI-NH3) 352 (M+
1
+ 1), 369 (M + NH4+); H NMR (CD3OD) δ 2.06 (3H, s, CH3-
2-(Eth ylth io)-N6-m eth yl-2′-d eoxya d en osin e (41). To a
suspension of potassium tert-butoxide (112 mg, 1 mmol) in dry
DMF (3 mL) in a sealed tube was added ethanethiol (60 mg,
71 µL, 1 mmol), and the reaction mixture was stirred at room
temperature for 30 min. 2-Chloro-2′-deoxy-N6-methyladenosine
(39; 50 mg, 0.17 mmol) in dry DMF (1 mL) was added and
heated at 110 °C for 4 h while stirring. The solvent was
removed under reduced pressure, and the residue was chro-
matographed using pTLC (CHCl3/MeOH, 9/1) to provide 41
as a colorless solid (39 mg, 0.12 mmol, 70% yield): MS (CI-
CO), 2.11 (3H, s, CH3CO), 2.53-2.60 (1H, m, CH2-2′), 2.96-
3.01 (1H, m, CH2-2′), 3.32 (1H, bs, H-4′), 4.25-4.36 (2H, m,
CH2-5′), 5.44 (1H, bs, H-3′), 6.27 (1H, t, J ) 6.4 Hz, H-1′), 7.87
(1H, s, H-8).
6-Ch lor o-9-(2-d eoxy-3,5-d i-O-a cetyl-â-D-er yth r op en to-
fu r a n osyl)gu a n in e (47). A flame-dried 50-mL round-bottom
flask fitted with a reflux condenser was charged with 46 (351
mg, 1 mmol), predried tetraethylammonium chloride (183 mg,
1.1 mmol), freshly distilled N,N-dimethylaniline (122 mg, 128
µL, 1 mmol), and dry acetonitrile (5 mL) under nitrogen atmos-
phere while stirring followed by cooling the mixture in an ice
1
NH3) 326 (M+ + 1); H NMR (CD3OD) δ 1.45 (3H, t, J ) 6.9
Hz, CH3CH2S), 2.45-2.52 (1H, m, CH2-2′), 2.88-297 (1H, m,

1636 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
bath. Phosphorus oxychloride (306 mg, 186 µL, 2 mmol) was
added dropwise, and stirring was continued for 10 min at room
temperature. The reaction mixture was then refluxed for 10
min. The solvent and volatile materials were evaporated under
reduced pressure, and the resultant yellow oil was stirred with
crushed ice for 15 min. The yellow oil was extracted with
chloroform (5 × 15 mL), washed with saturated NaHCO3
solution and cold water, dried over anhydrous Na2SO4, filtered,
and evaporated to give a light-yellow syrup which was purified
by flash column chromatography (CHCl3/MeOH, 95/5) to
furnish 47 as a light-yellow solid (164 mg, 0.44 mmol, 44%
yield): MS (CI-NH3) 370 (M+ + 1); 1H NMR (CDCl3) δ 2.11
(3H, s, CH3CO), 2.16 (3H, s, CH3CO), 2.67-2.74 (1H, m, CH2-
2′), 2.79-2.89 (1H, m, CH2-2′), 4.39 (3H, bs, CH2-5′ and H-4′),
5.40 (1H, m, H-3′), 6.45 (1H, t, J ) 6.9 Hz, H-1′), 8.34 (1H, s,
H-8).
2-Am in o-N6-m eth yl-2′-d eoxya d en osin e (48). A mixture
of 2-chloro-2′-deoxy-3′,5′-di-O-acetyl-6-chloropurine riboside
(47; 230 mg, 0.62 mmol) and 40% aqueous methylamine (5 mL)
was stirred at room temperature for 2 h. The solvent was
removed under reduced pressure, and the residue was chro-
matographed using pTLC (CHCl3/MeOH, 95/5) to obtain the
desired compound 2-chloro-2′-deoxy-N6-methyladenosine (48)
as a white solid (113 mg, 0.40 mmol, 55% yield): MS (CI-NH3)
281 (M+ + 1); 1H NMR (CD3OD) δ 2.28-2.33 (1H, m, CH2-2′),
2.77-2.82 (1H, m, CH2-2′), 3.04 (3H, s, N-CH3), 3.71-3.88
(2H, m, CH2-5′), 4.06 (1H, bs, H-4′), 4.56 (1H, bs, H-3′), 6.28
(1H, t, J ) 5.7 Hz, H-1′), 7.87 (1H, s, H-8).
pressure, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 9/1) to afford 52 as a colorless solid (25 mg,
1
0.08 mmol, 80% yield): MS (CI-NH3) 312 (M+ + 1); H NMR
(CD3OD) δ 1.44 (3H, t, J ) 7.8 Hz, CH3CH2S), 2.44-2.51 (1H,
m, CH2-2′), 2.90-2.98 (1H, m, CH2-2′), 3.24 (2H, q, J ) 6.86
Hz, CH3CH2S), 3.77-3.91 (2H, m, CH2-5′), 4.09 (1H, bs, H-4′),
4.62-4.64 (1H, m, H-3′), 6.46 (t, J ) 7.0 Hz, H-1′), 8.18 (1H,
s, H-8).
2-(P r op ylth io)-2′-d eoxya d en osin e (53). To a suspension
of potassium tert-butoxide (112 mg, 1 mmol) in dry DMF (2
mL) in a sealed tube was added propanethiol (75 mg, 89 µL, 1
mmol), and the reaction mixture was stirred at room temper-
ature for 30 min. A solution of 2-chloro-2′-deoxy-N6-methylad-
enosine (37; 31 mg, 0.1 mmol) in dry DMF (1 mL) was added
to the above mixture, and the reaction was stirred and heated
at 110 °C for 4 h. The solvent was removed under reduced
pressure, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 9/1) to afford 53 as a colorless solid (25 mg,
1
0.08 mmol, 76% yield): MS (CI-NH3) 326 (M+ + 1); H NMR
(CD3OD) δ 1.12 (3H, t, J ) 7.8 Hz, CH3CH2CH2S), 1.76-1.84
(2H, m, CH3CH2CH2S), 2.45-2.51 (1H, m, CH2-2′), 2.88-2.94
(1H, m, CH2-2′), 3.03 (2H, bs, 2OH), 3.18 (2H, q, J ) 5.9 Hz,
CH3CH2CH2S), 3.78-3.90 (2H, m, CH2-5′), 4.08-4.11 (1H, m,
H-4′), 4.62-4.65 (1H, m, H-3′), 6.43 (1H, t, J ) 6.8 Hz, H-1′),
8.03 (2H, bs, NH2), 8.22 (1H, s, H-8).
8-Br om o-2′-d eoxya d en osin e (55). A mixture of 2′-deoxy-
adenosine (54; 270 mg, 1 mmol) and N-bromosuccinimide (267
mg, 1.5 mmol) in dry DMF (6 mL) was stirred at room
temperature for 12 h under nitrogen atmosphere. The solvent
was removed under reduced pressure, and the residue was
purified using pTLC (CHCl3/MeOH, 9/1) to give 55 as a
colorless solid (132 mg, 0.40 mmol, 40% yield): MS (CI-NH3)
330 (M+ + 1); 1H NMR (CD3OD) δ 2.18-2.22 (1H, m, CH2-2′),
3.23-3.30 (1H, m, CH2-2′), 3.48-3.54 (1H, m, CH2-5′), 3.66-
3.70 (1H, m, CH2-5′), 3.92 (1H, bs, H-4′), 4.52 (1H, bs, H-3′),
5.31-5.38 (2H, bs, 2 OH), 6.32 (1H, t, J ) 6.9 Hz, H-1′), 7.55
(2H, bs, NH2), 8.14 (1H, s, H-2).
2-Am in o-N6-m et h yl-2′-d eoxy-3′,5′-d i-O-a cet yla d en os-
in e (49). A mixture of 2-amino-N6-methyl-2′-deoxyadenosine
(48, 140 mg, 0.5 mmol), pyridine (2 mL), and acetic anhydride
(1.5 mL) was stirred at room temperature for 4 h. The mixture
was evaporated to dryness and chromatographed by pTLC
(CHCl3/MeOH, 95/5) to obtain 49 as a colorless solid (142 mg,
1
0.39 mmol, 78% yield): MS (CI-NH3) 365 (M+ + 1); H NMR
(CDCl3) δ 2.09 (3H, s, CH3CO), 2.14 (3H, s. CH3CO), 2.59-
2.67 (1H, m, CH2-2′), 2.89-2.99 (1H, m, CH2-2′), 4.34-4.39
(3H, m, CH2-5′ and H-4′), 5.42-5.45 (1H, m, H-3′), 6.02 (2H,
bs, NH2), 6.44 (1H, t, J ) 5.9 Hz, H-1′), 8.24 (1H, s, H-8).
8-Met h oxy-2′-d eoxya d en osin e (56). To a solution of
8-bromo-2′-deoxyadenosine (55; 50 mg, 0.15 mmol) in dry DMF
(2 mL) was added sodium methoxide (54 mg, 1 mmol), and
the reaction mixture was stirred and heated in a sealed tube
at 110 °C for 6 h. The solvent was removed under reduced
pressure, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 9/1) to give 56 as a light-yellow solid (30 mg,
2-Iod o-N6-m eth yl-2′-d eoxya d en osin e (50). A flame-dried
25-mL round-bottom flask fitted with a reflux condenser was
charged with compound 49 (140 mg, 0.38 mmol), isoamyl
nitrite (1.5 mL) and diiodomethane (1.5 mL), under nitrogen
atmosphere. The mixture was stirred at 85 °C for 1 h. The
solvent was removed with nitrogen purging, and the crude
residue was stirred with aqueous methylamine (5 mL, 40%
solution) for 1 h. The water was removed under reduced
pressure, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 9/1) to furnish compound 50 as a colorless solid
(71 mg, 0.18 mmol, 48% yield): MS (CI-NH3) 392 (M+ + 1); 1H
NMR (CD3OD) δ 2.35-2.43 (1H, m, CH2-2′), 2.69-2.80 (1H,
m, CH2-2′), 3.06 (3H, s, NCH3), 3.71-3.82 (2H, m, CH2-5′),
4.02-4.05 (1H, m, H-4′), 4.54-4.58 (1H, m, H-3′), 6.34 (1H, t,
J ) 5.95 Hz, H-1′), 8.20 (1H, s, H-8).
1
0.11 mmol, 72% yield): MS (CI-NH3) 282 (M+ + 1); H NMR
(CD3OD) δ 2.20-2.27 (1H, m, CH2-2′), 2.48-2.62 (1H, m, CH2-
2′), 3.66 (3H, s, OCH3), 3.71-3.90 (2H, m, CH2-5′), 4.09 (1H,
bs, H-4′), 4.60 (1H, bs, H-3′), 6.34 (1H, t, J ) 6.0 Hz, H-1′),
8.06 (1H, s, H-2).
8-(Meth ylth io)-2′-d eoxya d en osin e (57). To a solution of
8-bromo-2′-deoxyadenosine (55, 50 mg, 0.15 mmol) in dry DMF
(2 mL) was added sodium thiomethoxide (70 mg, 1 mmol), and
the reaction mixture was stirred and heated in a sealed tube
at 110 °C for 4 h. The solvent was removed under reduced
pressure, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 9/1) to give the desired compound 8-(meth-
ylthio)-2′-deoxyadenosine (57) as a light-yellow solid (37.8 mg,
0.13 mmol, 85% yield): MS (EI) 297 (M+); 1H NMR (CD3OD) δ
2.20-2.28 (m, 1H, CH2-2′), 2.47-2.61 (1H, m, CH2-2′), 2.76
(3H, s, SCH3), 3.68-3.91 (2H, m, CH2-5′), 4.01 (1H, bs, H-4′),
4.60 (1H, m, H-3′), 6.36 (1H, t, J ) 5.9 Hz, H-1′), 8.10 (1H, s,
H-8).
2-(Meth ylth io)-2′-d eoxya d en osin e (51). To a solution of
2-chloro-2′-deoxyadenosine (37; 31 mg, 0.1 mmol) in dry DMF
(2 mL) was added sodium thiomethoxide (70 mg, 1 mmol), and
the mixture was stirred and heated in a sealed tube at 110 °C
for 4 h. The solvent was removed under reduced pressure, and
the residue was chromatographed using pTLC (CHCl3/MeOH,
93/7) to afford 51 as a white solid (21 mg, 0.07 mmol, 70%
yield): MS (CI-NH3) 298 (M+ + 1); 1H NMR (CD3OD) δ 2.28-
2.30 (1H, m, CH2-2′); 2.76-2.85 (1H, m, CH2-2′), 3.39 (3H, s,
CH3S), 3.52-3.68 (2H, m, CH2-5′), 3.91 (1H, m, H-4′), 4.47 (1H,
m, H-3′), 4.99 (1H, bs, OH), 5,38 (1H, bs, OH), 6.35 (1H, t, J )
5.7 Hz, H-1′), 7.41 (2H, bs, NH2), 8.27 (1H, s, H-8).
2-(Eth ylth io)-2′-d eoxya d en osin e (52). To a suspension
of potassium tert-butoxide (112 mg, 1 mmol) in dry DMF (2
mL) in a sealed tube was added ethanethiol (62 mg, 74 µL, 1
mmol), and the reaction mixture was stirred at room temper-
ature for 30 min. A solution of 2-chloro-2′-deoxy-N6-methylad-
enosine (37; 31 mg, 0.1 mmol) in dry DMF (1 mL) was added
to the above mixture, and the reaction was stirred and heated
at 110 °C for 4 h. The solvent was removed under reduced
8-(Meth yla m in o)-2′-d eoxya d en osin e (58). A 40% aque-
ous solution of methylamine (5 mL) was added to 8-bromo-2′-
deoxyadenosine (55, 50 mg, 0.15 mmol), and the reaction
mixture was stirred at room temperature for 3 h. The solvent
was removed under nitrogen flow, and the residue was purified
using pTLC (CHCl3/MeOH, 9/1) to afford 58 as a white solid
(37.8 mg, 0.14 mmol, 90% yield): MS (CI-NH3) 281 (M+ + 1);
1H NMR (CD3OD) δ 2.11-2.52 (2H, m, CH2-2′), 3.31 (3H, s,
NCH3), 3.81-3.84 (2H, m, CH2-5′), 4.06 (1H, m, H-4′), 4.58 (1H,
m, H-3′), 6.44 (1H, t. J ) 5.8 Hz, H1′), 7.99 (1H, s, H-8).
2′-Deoxy-N6-m eth oxyla d en osin e (59). To a mixture of
6-chloro-2′-deoxypurine riboside (25 mg, 0.092 mmol) and

Bisphosphates as P2Y1 Receptor Antagonists
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9 1637
methoxyamine hydrochloride (100 mg, 1.19 mmol) in DMF (2
mL) was added 1 M NaOH (1.19 mL). The reaction mixture
was stirred at room temperature for 72 h and then neutralized
with 0.1 M HCl. The mixture was evaporated to dryness and
chromatographed by pTLC (CHCl3/MeOH, 90/10) to obtain the
desired compound 59 (14 mg, 0.050 mmol, 54% yield): MS (CI-
NH3) 282 (M+ + 1), 299 (M + NH4+); 1H NMR (CD3OD) δ 2.52-
2.58 (1H, m, CH2-2′), 2.74-2.83 (1H, m, CH2-2′), 3.80 (2H, m,
CH2-5′), 3.85 (3H, s, OCH3), 4.14 (1H, bs, H-4′), 4.61 (1H, m,
H-3′), 6.39 (1H, t, J ) 6.9 Hz, H-1′), 7.93 (1H, s, H-2), 8.14
(1H, s, H-8).
Car bocyclic 2-Am in o-N6-m eth yl-2′-deoxyaden osin e (61).
A solution of carbocyclic 2-amino-6-chloro-2′-deoxyadenosine
(60; 28 mg, 0.1 mmol) and 40% aqueous methylamine (4 mL)
was stirred at room temperature for 3 h. The solvent was
removed under reduced pressure, and the residue was chro-
matographed using pTLC (CHCl3/MeOH, 9/1) to provide 61
as colorless solid (26 mg, 0.09 mmol, 92% yield): MS (CI-NH3)
279 (M+ + 1); 1H NMR (CD3OD) δ 1.72-1.79 (1H, m, CH2-2′),
2.01-2.19 (2H, m, CH2-6′), 2.20-2.29 (1H, m, CH2-2′), 2.32-
2.43 (1H, m, H-1′), 2.96 (3H, s, N-CH3), 3.58-3.69 (2H, m,
CH2-5′), 4.20 (1H, bs, H-3′), 7.75 (1H, s, H-8).
Stimulation of inositol phosphate formation in 1321N1
human astrocytoma cells stably expressing recombinant hu-
man P2Y2 receptors (activated by 100 nM UTP), recombinant
human P2Y4 receptors (activated by 100 nM UTP), and
recombinant rat P2Y6 receptors (activated by 100 nM UDP)
was measured in a similar fashion.21,38
Da ta An a lysis. Agonist potencies were calculated using a
four-parameter logistic equation and the GraphPad software
package (GraphPad, San Diego, CA). EC50 values (mean (
standard error) represent the concentration at which 50% of
the maximal effect is achieved. Relative efficacies (%) were
determined by comparison with the effect produced by a
maximal effective concentration of 2-MeSADP in the same
experiment.
Antagonist IC50 values (mean ( standard error) represent
the concentration needed to inhibit by 50% the effect elicited
by 10 nM 2-MeSADP. The percent of maximal inhibition is
equal to 100 minus the residual fraction of stimulation at the
highest antagonist concentration.
All concentration-effect curves were repeated in at least
three separate experiments carried out with different mem-
brane preparations using duplicate or triplicate assays.
2′-Deoxy-4′-th ioa d en osin e S-Oxid e (63). A solution of
sodium metaperiodate (22.5 mg, 0.105 mmol) in water (0.4 mL)
was added dropwise to a vigorously stirred ice-cold solution
of 2′-deoxy-4′-thioadenosine (62; 13.3 mg, 0.050 mmol) in
methanol (1 mL). The reaction mixture was stirred for 48 h
at room temperature. The solvent was removed under nitrogen
purging, and the residue was chromatographed using pTLC
(CHCl3/MeOH, 10/1) to obtain the desired 2′-deoxy-4′-thioad-
enosine S-oxide, 63 (mixture of diastereomers), as a colorless
solid (14.0 mg, 0.049 mmol, 98% yield): MS (CI-NH3) 284 (M+
Abbr evia tion s: ATP, adenosine 5′-triphosphate; DEAE,
diethylaminoethyl; DMF, dimethylformamide; DMSO, dim-
ethyl sulfoxide; FAB, fast atom bombardment (mass spectros-
copy); HPLC, high-pressure liquid chromatography; MS, mass
spectroscopy; HRMS, high-resolution mass spectroscopy; IAN,
isoamyl nitrite; 2-MeSADP, 2-(methylthio)adenosine 5′-diphos-
phate; NBS, N-bromosuccinimide; NNDA, N,N-dimethyla-
niline; Py, pyridine; TBAP, tetrabutylammonium phosphate;
TEAA, triethylammonium acetate; pTLC, preparative thin-
layer chromatography.
1
+ 1); H NMR (CD3OD) δ 2.70-3.19 (2H, m, CH2-2′), 3.44-
3.48 (1H, m, H-4′), 4.01-4.07 (2H, m, CH2-5′), 4.55-4.66 (1H,
2m, H-3′), 6.08-6.28 (1H, 2m, H-1′), 8.20, 8.25 (1H, 2s, H-2),
8.26, 8.29 (1H, 2s, H-8).
Ack n ow led gm en t. We thank Gilead Sciences (Fos-
ter City, CA) for financial support to E.N. and Dr.
Stefano Moro (NIDDK) for helpful discussions. We also
are grateful to Dr. Lewis Pannell, Noel Whittaker, Mary
Furr, and Wesley White for technical assistance. This
work was supported by USPHS Grants GM38213 and
HL54889.
2′-De oxy-5′-[d ie t h yl(O-p h osp h on ylm e t h yl)]a d e n os-
in e (64). A flame-dried 50-mL round-bottom flask was charged
with 2′-deoxyadenosine (51; 270 mg, 1 mmol) and potassium
tert-butoxide (146 mg, 1.3 mmol) in dry DMF (5 mL) under
argon atmosphere, and the reaction mixture was stirred at
room temperature for 2 h. A solution of diethyl tosylmetha-
nephosphonate (320 mg, 1.1 mmol) in dry DMF (1 mL) was
added dropwise, and the reaction mixture was allowed to stir
for 3 days at room temperature. The solvent was removed by
nitrogen purging, and the crude product was purified by pTLC
(CHCl3/MeOH, 85/15) to afford 64 as a light-yellow solid (128
mg, 0.32 mmol, 32% yield): MS (CI-NH3) 402 (M+ + 1); 1H
NMR (CD3OD) δ 1.21-1.25 (6H, m, OCH2CH3); 2.34-2.42 (1H,
m, CH2-2′), 2.74-2.85 (1H, m, CH2-2′), 3.44 (2H, d, J ) 10.5
Hz, -CH2-P), 3.71-4.13 (6H, m, CH2-5′ and P-CH2CH3), 4.19
(1H, m, H-4′), 4.60 (1H, m, H-3′), 6.42 (1H, t, J ) 6.5 Hz, H-1′),
7.97 (2H, bs, NH2), 8.18 (1H, s, H-2), 8.26 (1H, s, H-8).
P h ar m acological An alyses. P2Y1 receptor-promoted stimu-
lation of inositol phosphate formation by adenine nucleotide
analogues was measured in turkey erythrocyte membranes as
previously described.10,39 The EC50 values were averaged from
3-8 independently determined concentration-effect curves for
each compound. Briefly, 1 mL of washed turkey erythrocytes
was incubated in inositol-free medium (DMEM; Gibco, Gaith-
ersburg, MD) with 0.5 mCi of 2-[3H]myo-inositol (20 Ci/mmol;
American Radiolabeled Chemicals, Inc., St. Louis, MO) for 18-
24 h in a humidified atmosphere of 95% air/5% CO2 at 37 °C.
Erythrocyte ghosts were prepared by rapid lysis in hypotonic
buffer (5 mM sodium phosphate, pH 7.4, 5 mM MgCl2, 1 mM
EGTA) as described.39 Phospholipase C activity was measured
in 25 µL of [3H]inositol-labeled ghosts (approximately 175 µg
of protein, 200-500 000 cpm/assay) in a medium containing
424 µM CaCl2, 0.91 mM MgSO4, 2 mM EGTA, 115 mM KCl, 5
mM KH2PO4, and 10 mM Hepes, pH 7.0. Assays (200 µL final
volume) contained 1 µM GTPγS and the indicated concentra-
tions of nucleotide analogues. Ghosts were incubated at 30 °C
for 5 min, and total [3H]inositol phosphates were quantitated
by anion-exchange chromatography as previously described.10,39
Refer en ces
(1) Fredholm, B. B.; Abbracchio, M. P.; Burnstock, G.; Dubyak, G.
R.; Harden, T. K.; J acobson, K. A.; Schwabe, U.; Williams, M.
Toward a revised nomenclature for P1 and P2 receptors. Trends
Pharm. Sci. 1997, 18, 79-82.
(2) North, R. A.; Barnard, E. A. Nucleotide receptors. Curr. Opin.
NeuroBiol. 1997, 7, 346-357.
(3) Webb, T. E.; Simon, J .; Krishek, B. J .; Bateson, A. N.; Smart, T.
G.; King, B. F.; Burnstock, G.; Barnard, E. A. Cloning and
functional expression of a brain G-protein-coupled ATP receptor.
FEBS Lett. 1993, 324, 219-225.
(4) Schachter, J . B.; Li, Q.; Boyer, J . L.; Nicholas, R. A.; Harden, T.
K. 2nd messenger cascade specificity and pharmacological
selectivity of the human P2Y1-purinoceptor. Br. J . Pharmacol.
1996, 118, 167-173.
(5) J anssens, R.; Communi, D.; Pirotton, S.; Samson, M.; Parmen-
tier, M.; Boeynaems, J .-M. Cloning and tissue distribution of the
human P2Y1 receptor. Biochem. Biophys. Res. Commun. 1996,
221, 588-593.
(6) Ayyanathan, K.; Webb, T. E.; Sandhu, A. K.; Athwal, R. S.;
Barnard, E. A.; Kunapuli, S. P. Cloning and chromosomal
localization of the human P2Y1 purinoceptor. Biochem. Biophys.
Res. Commun. 1996, 218, 783-788.
(7) Burnstock, G.; King, B. F. Numbering of cloned P2 purinoceptors.
Drug Dev. Res. 1996, 38, 67-71.
(8) Communi, D.; Govaerts, C.; Parmentier, M.; Boeynaems, J . M.
Cloning of a human purinergic P2Y receptor coupled to phos-
pholipase C and adenylyl cyclase. J . Biol. Chem. 1997, 272,
31969-31973.
(9) Lazarowski, E. R.; Homolya, L.; Boucher, R. C.; Harden, T. K.
Direct demonstration of mechanically induced release of cellular
UTP and its implication for uridine nucleotide receptor activa-
tion. J . Biol. Chem. 1997, 272, 24348-24354.
(10) Harden, T. K.; Hawkins, P. T.; Stephens, L.; Boyer, J . L.;
Downes, P. Phosphoinositide hydrolysis by guanosine 5′-(gamma-
thio]triphosphate-activated phospholipase C of turkey erythro-
cyte membranes. Biochem. J . 1988, 252, 583-593.

1638 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 9
Nandanan et al.
(11) J in, J .; Daniel, J . L.; Kunapuli, S. P. Molecular basis for ADP-
induced platelet activation. II. The P2Y1 receptor mediates ADP-
induced intracellular calcium mobilization and shape change in
platelets. J . Biol. Chem. 1998, 273, 2030-2034.
(12) 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.
(13) Crowley, M. R. Oxygen-induced pulmonary vasodilation is
mediated by adenosine triphosphate in newborn lambs. J .
Cardiovasc. Pharmacol. 1997, 30, 102-109.
(14) Loubatie`res-Mariani, M.-M.; Hillaire-Buys, D.; Chapal, J .; Ber-
trand, G.; Petit, P. P2 purinoceptor agonists: New insulin
secretagogues potentially useful in the treatment of noninsulin-
dependent diabetes mellitus. In Purinergic Approaches in Ex-
perimental Therapeutics; J acobson, K. A., J arvis, M. F., Eds.;
Wiley: New York, 1997; pp 253-260.
(15) Fagura, M. S.; Dainty, I. A.; McKay, G. D.; Kirk I. P.; Humphries,
R. G.; Robertson, M. J .; Dougall, I. G.; Leff, P. P2Y1-receptors
in human platelets which are pharmacologically distinct from
P2Y(ADP)-receptors. Br. J . Pharmacol. 1998, 124, 157-164.
(16) Virginio, C.; Robertson, G.; Surprenant, A.; North, R. A. Trini-
trophenyl-substituted nucleotides are potent antagonists selec-
tive for P2X1, P2X3, and heteromeric P2X2/3 receptors. Mol.
Pharmacol. 1998, 53, 969-973.
(17) Humphreys, B. D.; Virginio, C.; Surprenant, A.; Rice, J .; Dubyak,
G. R. Isoquinolines as antagonists of the P2X7 nucleotide
receptor: high selectivity for the human versus rat receptor
homologues. Mol. Pharmacol. 1998, 54, 22-32.
(18) Kim, Y.-C.; Camaioni, E.; Ziganshin, A. U.; J i, X.-J .; King, B.
F.; Wildman, S. S.; Rychkov, A.; Yoburn, J .; Kim, H.; Mohanram,
A.; Harden, T. K.; Boyer, J . L.; Burnstock, G.; J acobson, K. A.
Synthesis and structure activity relationships of pyridoxal-6-
azoaryl-5′-phosphate and phosphonate derivatives as P2 receptor
antagonists. Drug Dev. Res., in press.
(19) Boyer, J . L.; Romero-Avila, T.; Schachter, J . B.; Harden, T. K.
Identification of competitive antagonists of the P2Y1 receptor.
Mol. Pharmacol. 1996, 50, 1323-1329.
(20) 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.
(21) Boyer, J . L.; Mohanram, A.; Camaioni, E.; J acobson, K. A.;
Harden, T. K. Competitive and selective antagonism of P2Y1
receptors by N6-methyl 2′-deoxyadenosine 3′,5′-bisphosphate.
Brit. J . Pharmacol. 1998, 124, 1-3.
(22) 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.
(23) 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; Abstact A03.
(24) Cristalli, G.; Vittori, S.; Eleuteri, A.; Volpini, R.; Camaioni, E.;
Lupidi, G. Synthesis of 2′-deoxynucleoside derivatives of 1-dea-
zapurine. Nucleosides Nucleotides 1994, 13, 835-848.
(25) Verheggen, I.; van Aerschot, A.; Toppet, S.; Snoeck, R.; J anssen,
G.; Balzarini, J .; de Clercq, E.; Herdewijn, P. Synthesis and
antiherpes virus activity of 1,5-anhydrohexitol nucleosides. J .
Med. Chem. 1993, 36, 2033-2040.
(26) Messini, L.; Tiwari, K. N.; Montgomery, J . A.; Secrist, J . A. III.
Synthesis and Biological Activity of 4′-Thio-2′-deoxy Purine
Nucleosides; 13th International Round Table on Nucleosides,
Nucleotides and their Biological Applications; Sept. 6-10, 1998.
(27) Shealy, Y. F.; O’Dell, C. A. Carbocyclic Analogues of 2′-Deoxy-
adenosine. Tetrahedron Lett. 1969, 2231-2234.
(28) Robins, M. J .; Fouron, Y.; Mengel, R. Adenosine 2′,3′-ribo-
epoxide. Synthesis, intramolecular degradation, and transforma-
tion into 3′-substituted xylofuranosyl nucleosides and the lyxo-
epoxide. J . Org. Chem. 1974, 39, 1564-1570.
(29) Robins, M. J .; Uza’nski, B. Nucleic acid related compounds. 33.
Conversions of adenosine and guanosine to 2,6-dichloro, 2-amino-
6-chloro, and derived nucleosides. Can. J . Chem. 1981, 59, 2601-
2607.
(30) van Aerschot, A.; Mamos, P.; Weyns, N.; Ikeda, S.; de Clercq,
E.; Herdewijn, P. Antiviral activity of C-alkylated purine nucleo-
sides obtained by cross-coupling with tetraalkyltin reagents. J .
Med. Chem. 1993, 36, 2938-2942.
(31) Hoffman, D. J .; Whistler, R. L. Synthesis and properties of
nucleotides containing 4-thio-D-ribofuranoside. Biochemistry
1970, 9, 2367-2372.
(32) Hancox, E. L.; Walker, R. T. Some reactions of 4′-thionucleosides
and their sulfones. Nucleosides Nucleotides 1996, 15, 1-3.
(33) 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.
(34) Boyer, J . L.; Siddiqi, S.; Fischer, B.; Romero-Avila, T.; J acobson,
K. A.; Harden, T. K. Identification of potent P2Y-purinoceptor
agonists that are derivatives of adenosine 5′-monophosphate. Br.
J . Pharmacol. 1996, 118, 1959-1964.
(35) van der Wenden, E. M.; Carnielli, M.; Roelen, H. C.; Lorenzen,
A.; von Frijtag Drabbe Ku¨nzel, J . K.; IJ zerman, A. P. 5′-
substituted adenosine analogues as new high-affinity partial
agonists for the adenosine A1 receptor. J . Med. Chem. 1998, 41,
102-108.
(36) J asper, J . R.; Insel, P. A. Evolving concepts of partial agonism.
The beta-adrenergic receptor as a paradigm. Biochem. Pharma-
col. 1992, 43, 119-130.
(37) J acobson, K. A.; Hoffmann, C.; Kim, Y.-C.; Camaioni, E.;
Nandanan, E.; J ang, S. Y.; Guo, D.-P.; J i, X.-d.; von Ku¨gelgen,
I.; Moro, S.; Ziganshin, A. U.; Rychkov, A.; King, B. F.; Brown,
S.; Wildman, S. S.; Burnstock, G.; Boyer, J . L.; Mohanram, A.;
Harden, T. K. Molecular recognition in P2 receptors: Ligand
development aided by molecular modeling and mutagenesis.
Prog. Brain Res., in press.
(38) J acobson, K. A.; Kim, Y.-C.; Wildman, S. S.; Mohanram, A.;
Harden, T. K.; Boyer, J . L.; King, B. F.; Burnstock, G. A
pyridoxine cyclic-phosphate and its 6-arylazo-derivative selec-
tively potentiate and antagonize activation of P2X1 receptors.
J . Med. Chem. 1998, 41, 2201-2206.
(39) Boyer, J . L.; Downes, C. P.; Harden, T. K. Kinetics of activation
of phospholipase C by P2Y purinergic receptor agonists and
guanine nucleotides. J . Biol. Chem. 1989, 264, 884-890.
J M980657J