Lipophilic modifications to dinucleoside polyphosphates and nucleotides that confer antagonist properties at the platelet P2Y12 receptor

J. Med. Chem. 2008, 51, 1007–1025

1007

Lipophilic Modiﬁcations to Dinucleoside Polyphosphates and Nucleotides that Confer

Antagonist Properties at the Platelet P2Y₁₂Receptor

James G. Douglass,*^,†Roshni I. Patel,^‡Benjamin R. Yerxa,^†Sammy R. Shaver,^†Paul S. Watson,^†Krzysztof Bednarski,^†

Robert Plourde,^†Catherine C. Redick,^‡Kurt Brubaker,^‡Arthur C. Jones,^‡and José L. Boyer^‡

Departments of Chemistry and Molecular Pharmacology, Inspire Pharmaceuticals, Inc., 4222 Emperor BouleVard, Suite 200,

Durham, North Carolina 27703-8466

ReceiVed October 26, 2007

Platelet P2Y₁₂receptors play a central role in the regulation of platelet function and inhibition of this receptor

by treatment with drugs such as clopidogrel results in a reduction of atherothrombotic events. We discovered

that modiﬁcation of natural and synthetic dinucleoside polyphosphates and nucleotides with lipophilic

substituents on the ribose and base conferred P2Y₁₂receptor antagonist properties to these molecules producing

potent inhibitors of ADP-mediated platelet aggregation. We describe methods for the preparation of these

functionalized dinucleoside polyphosphates and nucleotides and report their associated activities. By analysis

of these results and by deconstruction of the necessary structural elements through selected syntheses, we

prepared a series of highly functionalized nucleotides, resulting in the selection of an adenosine monophosphate

derivative (62) for further clinical development.

Introduction

blood clots. While this function is vital in the maintenance of

hemostasis, platelet activation under various pathological condi-

tions can potentially lead to life-threatening thrombotic events.

Extracellular nucleotides and dinucleoside polyphosphates

participate in the regulation of a broad range of physiological

functions in numerous cell types and tissues, acting as agonists

or antagonists on P2 cell surface receptors.^1–3The P2 receptor

superfamily is composed of two subfamilies designated P2X

and P2Y, each having multiple receptor subtypes. The P2X

subfamily, composed of ligand-gated cation channels, has seven

subtypes (P2X₁-P2X₇). The P2Y subfamily, composed of

G-protein-coupled receptors has eight subtypes (P2Y₁, P2Y₂,

P2Y₄, P2Y₆, P2Y₁₁, P2Y₁₂, P2Y₁₃, and P2Y₁₄). In addition to

its ubiquitous role in cellular energy pathways, adenosine 5′-

triphosphate (ATP, 3, Figure 1) is the endogenous agonist of

the seven P2X receptor subtypes as well as the P2Y₂and P2Y₁₁

receptors. Adenosine 5′-diphosphate (ADP, 2) is the natural

agonist of P2Y₁, P2Y₁₂, and P2Y₁₃receptors. The pyrimidine

nucleotide uridine 5′-triphosphate (UTP, 6) is also an agonist

of P2Y₂, with potency comparable to that of ATP. UTP also

serves as the ligand for the P2Y₄receptor, while its correspond-

ing diphosphate, UDP (5), is the agonist of the P2Y₆receptor.

UDP-glucose is the agonist of the most recently described

member of the P2Y subfamily, the P2Y₁₄receptor.⁴Dinucleo-

side polyphosphates are also endogenous substances released

into in the extracellular space, where they are known to stimulate

several P2 receptors with activities comparable to that of their

nucleotide counterparts. Natural and synthetic dinucleoside

polyphosphates have agonist potencies comparable to those of

ATP and UTP at P2Y₂receptors and to that of UDP at the P2Y₆

receptor. The inherently greater stability of dinucleoside poly-

phosphates compared to that of their corresponding nucleotides

has led us to develop two different P2Y₂dinucleoside tetra-

phosphate agonists (13 and 14) for the treatment of dry eye

disease and cystic ﬁbrosis.^5,6

Platelets express a variety of cell-surface receptors, three of

which belong to the P2 superfamily (P2X₁, P2Y₁, and P2Y₁₂).

The role of P2X₁on platelet function is not completely

understood; it has been demonstrated recently using P2X₁

receptor knockout mice that this receptor plays an important

role in thrombus formation under high shear conditions, as might

be found in a partially occluded vessel.⁷In contrast, the

participation of P2Y₁and P2Y₁₂receptors in platelet activation

and aggregation have been studied extensively.^8,9,10

Activation of P2Y₁by ADP results in platelet shape change

and transient aggregation, whereas activation of P2Y₁₂leads to

degranulation and release of more endogenous nucleotides,

leading to sustained aggregation. ADP is found in high

concentrations in platelet granules and is released in response

to stimulation with a number of platelet-aggregating agents and

by the exposure of platelets to foreign surfaces or to the pro-

aggregatory subendothelium following the injury of the blood

vessel wall. Activated platelets also produce thromboxane A2

(TXA₂) and release ﬁbrinogen and other active substances from

storage granules. TXA₂ampliﬁes shape change and leads to

further activation of the platelet. Platelet activation and de-

granulation leads to further recruitment of platelets to the site

of injury. Ultimately, ﬁbrinogen cross-links platelets via the

glycoprotein IIb/IIIa receptor, leading to the formation of a stable

thrombus.

ADP-promoted aggregation requires the simultaneous activa-

tion of both the P2Y₁and P2Y₁₂receptors, and inhibition of

ADP binding at either receptor is sufﬁcient to prevent platelet

aggregation. The P2Y₁₂receptor is found primarily on platelets

and in the central nervous system, whereas the P2Y₁receptor

has a more widespread distribution.⁴The comparatively limited

distribution of P2Y₁₂receptors, coupled to the fact that this

receptor has been validated by the thienopyridines as a clinical

target, makes the P2Y₁₂receptor an attractive drug discovery

target for the development of medicines to treat cardiovascular

diseases.^11,12

Platelets are small disk-shaped anucleated cells found in

circulation, which are responsible for initiating and stabilizing

* To whom correspondence should be addressed. Tel: (919) 941-9777.

Fax: (919) 941-9177. E-mail: jdouglass@inspirepharm.com.

†

Department of Chemistry.

Department of Molecular Pharmacology.

‡

10.1021/jm701348d CCC: $40.75

2008 American Chemical Society

Published on Web 01/31/2008

1008 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

Figure 1. Structures of selected compounds.

In addition to its role as the agonist of multiple P2 receptors,

ATP also antagonizes the action of ADP at the platelet P2Y₁

and P2Y₁₂receptors.^13,14Substitutions at the 2 and 6 positions

of ATP have been shown to amplify both the potency and

selectivity of this antagonist (see Figure 1 for ring numbering).

The ATP analogue AR-C69931MX (cangrelor, 7) is a highly

potent P2Y₁₂antagonist representative of this molecular class

and is currently being evaluated in clinical trials for use during

percutaneous coronary interventions.^15,16Further work by the

Astra Zeneca group led to the development of AZD6140 (8), a

heavily modiﬁed adenosine-like molecule being evaluated as

an oral drug for chronic therapy for the prevention of thrombotic

events in patients with acute coronary syndrome.¹⁷

Several other molecular chemotypes have been shown to

reversibly inhibit ADP-induced platelet aggregation. Scarbor-

ough et al.¹⁸reported on a series of benzothiazolo[2,3-

c]thiadiazines and dioxo quinazolin sulfonyl ureas, which show

high potency in platelet aggregation and P2Y₁₂receptor binding

assays. Recently, Wang et al.¹⁹described results for a quinolin/

piperazin 5-oxopentanoic acid derivative, which showed an

improved therapeutic index over clopidogrel in animal models.

The clinical use of the thienopyridines clopidogrel (9) and

ticlopidine (10) has been shown to reduce the incidence of

myocardial infarction and stroke in patients with a history of

vascular disease and to provide protection from adverse events

following stent implantation.^20–22However, these drugs bind

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1009

to the P2Y₁₂receptor in an irreversible manner following

conversion to their respective active species by hepatic me-

tabolism.²³The irreversible nature of the inhibition of platelet

function by clopidogrel and ticlopidine preclude their use in

acute cardiovascular and cerebrovascular conditions where the

strict control of platelet function is needed. Furthermore,

signiﬁcant interindividual variability in response to the drug has

been seen in patients treated with clopidogrel.²⁴Treatment with

thienopyridines has also been associated with thrombotic

thrombocytopenic purpura, a rare but potentially fatal side

effect.²⁵More recently, prasugrel, a third generation thienopy-

ridine with apparently improved pharmacokinetic properties, has

been developed.^26,27Nevertheless, in spite of apparent progress

in the thienopyridine class of irreversible inhibitors, opportunities

still remain for the development of safe, rapidly reversible

platelet aggregation inhibitors for the treatment of both acute

and chronic clinical conditions.

Dinucleoside polyphosphates (Np_nN′), particularly those

having adenine as the base, have been detected in chromafﬁn

cells and platelet granules.^28,29Physiological roles of diadenosine

polyphosphates include inhibition of adenosine kinase,³⁰release

of nitric oxide from epithelial cells,³¹neurotransmitter release

at synaptic junctions,³²and inhibition of platelet aggregation.^33,34

In addition to being a potent P2Y₂agonist, Ap₄A (11) is a

modestly potent antagonist of ADP-induced platelet aggregation.

Several analogues of Ap₄A with modiﬁcations in the phosphate

chain (e.g., 12)³⁵were prepared in an effort to improve their

biological stability and to examine their potency for the

inhibition of platelet aggregation.

ined. Previously, it was observed that acetal modiﬁcations

to an analogue of 3 resulted in compounds that showed

activity at several P2X receptors.³⁷

A general method of forming cyclic acetals of acid

sensitive dinucleoside polyphosphates and nucleotides was

developed for the purpose of this work. Modest yielding

reactions (see Table 1) were achieved with most substrates

using the sodium salts of the starting dinucleoside polyphos-

phates or nucleotides with either 99% formic acid or

anhydrous triﬂuoroacetic acid as both the catalyst and solvent

in conjunction with 2–4 equivalents of the aldehyde or

aldehyde dimethylacetal.

Treatment of symmetrical dinucleoside tri and tetraphosphates

11, 13, and 19 with a variety of aldehydes in acid gave a mixture

of mono and diacetals (37–38, 20–31, and 35–36, respectively;

formulas II and IV, Scheme 1). Likewise, similar treatment of

the unsymmetrical dinucleoside tetraphosphate 18 gave the two

possible regioisomeric monoacetals (32 and 33; formulas II and

III) and diacetal 34 (formula IV). The structures of these

compounds and their biological evaluation as antagonists of

ADP-induced aggregation in a washed platelet assay are

presented in Table 1.

Dinucleoside tetraphosphates 42 and 47–50 were prepared

via an indirect synthetic strategy that also provided access to a

variety of substituted nucleotide derivatives. This synthetic

procedure gave better results than the direct action of an

isocyanate on diacetals 38 and 34.

Because of our long-standing interest in the pharmacologi-

cal properties and structure–activity relationships (SAR) of

dinucleoside polyphosphates as agonists of the P2 superfamily

of receptors, particularly P2Y₂, we synthesized a number of

analogues bearing lipophilic modiﬁcations to the ribose and

the base. During the course of this work, we found that

esteriﬁcation of one or more of the 2′- or 3′-ribose hydroxyls

of a P2Y₂receptor agonist produced compounds that

antagonized the ability of ADP to induce platelet aggregation.

In this article, we describe the identiﬁcation and SAR of

substitutions to dinucleoside polyphosphates and nucleotides

that confer antagonist properties for the P2Y₁₂receptor.

These alternative strategies are outlined in Schemes 2 and

3. For the adenosine analogues, ADP (2) was converted to

acetal 40 with phenylacetaldehyde dimethyl acetal in formic

acid and transformed to the urea with phenylisocyanate in

DMF to give 41. Carbonyldiimidazole activation of 41

facilitated self-condensation to give 42. To evaluate the effect

of the urea modiﬁcation in the platelet aggregation assay, 2

was also directly converted to the urea with phenylisocyanate

to give 39.³⁸For the cytidine analogues, cytidine monophos-

phate (43) was converted to urea with p-ﬂuorophenylisocy-

anate in DMF to give 44. Use of 44 in the dinucleoside

polyphosphate-forming reaction gave a better outcome than

43. UTP (6) was activated with dicyclohexylcarbodiimide,

leading to the cyclical trimeta-phosphate,³⁹and was con-

densed with 44 to give 47. Treatment of 47 with phenylac-

etaldehyde dimethylacetal in formic acid gave monoacetals

48 and 49 and diacetal 50. For purposes of biological

comparison, the same acetal and urea modiﬁcations were

made to 43, to give nucleotides 45 and 46. Similarly, AMP

(1) was converted to its corresponding 2′,3′-acetal

(Scheme 4) providing nucleotides 52–55. Compounds 52–55

were further modiﬁed to urea moieties at the 6-position of

the purine by reaction of the amino group with an appropriate

isocyanate in DMF. For these substrates, the isocyanate also

acted as a weak condensing agent for the phosphate group,

leading to variable amounts of the corresponding dinucleoside

diphosphates as a byproduct (Scheme 4). These byproducts

were easily separated from the desired nucleotides (56–66)

during puriﬁcation by HPLC. Biological results for these

analogues are presented in Table 2. The biological results

for compounds derived from 1 are presented in Table 3.

Chemistry

Ready access to commercially available nucleotides and

an in-house collection of dinucleoside polyphosphates led

us to choose these molecules as starting materials for our

syntheses. To streamline the synthesis of the required

analogues, the 5′-phosphate group was used as a protecting

group. This strategy eliminated the long protection/depro-

tection procedures that have traditionally been the norm in

nucleoside chemistry and facilitated the use of preparative

reverse-phase HPLC to purify these water soluble but

lipophilic products.

Starting with 15, 16, and 17 (Figure 1) as inspiration, we

considered alternative modiﬁcations to the ribose hydroxyl

groups. In addition to not being particularly stable, the

monoesters of 2′, 3′-ribose diols demonstrate no chemose-

lectivity toward either the 2′ or 3′ group, providing mixtures

of products upon reaction with acylating agents.³⁶Further-

more, mixtures of products often result upon subsequent

separation and puriﬁcation because of the migration of the

ester moiety between the adjacent hydroxyl groups. To

achieve greater stability and access to regiochemically

discreet analogues, cyclic acetals were prepared and exam-

Acetal-forming reactions using benzaldehyde, cinnamal-

dehyde, and phenylpropargyl aldehyde produced mixtures of

1010 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

Table 1. Inhibition of Platelet Aggregation and Selectivity for Dinucleoside Tri and Tetraphosphate Acetal Analogues^a

isolated

conversion yield^d

platelet aggregation

compd formula B₁B₂

R

n

(%)

IC₅₀(µM) mean ( SEM

P2Y₁(EC₅₀)

P2Y₂(EC₅₀)

P2Y₆(EC₅₀)

11

13

18

19

15

16

17

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

I

A

U

A

U

C

U

C

U

A

NA

3

2

3

2

3

ND

41

21

52

18

42

24

37

9

40

50

31

63

22

ND

31

33.5

31

30

18

42

6.7

38

9.7

32

3.5

23

30

25

32

20

25

19

38

63% inhibition at 100 µM 0.426 ( 0.236 0.072 ( 0.022 NR

NR

SR

NR

0.034 ( 0.011 18.64 ( 6.11

0.044 ( 0.004 SR

13.54 ( 5.71

0.40 ( 0.258

NA

II

IV

II

IV

II

IV

II

IV

II

IV

II

IV

II

III

IV

II

46% inhibition at 100 µM NR

7.59 ( 0.42 NR

52% inhibition at 100 µM NR

0.130 ( 0.031 SR

5.47 ( 1.16

NR

SR

NR

benzyl

27.1 ( 6.64

1.82 ( 0.717

16.7 ( 3.45

8.29 ( 1.7

8.8 ( 3.96

5.51 ( 2.12

8.68 ( 2.01

3.3 ( 1.98

NR

0.094 ( 0.033 SR

14.6 ( 0.94 SR

0.192 ( 0.052 SR

3.07 ( 2.15 NR

0.155 ( 0.116 NA

NR SR

0.068 ( 0.019 NA

0.35 ( 0.134 NR

0.098 ( 0.037 30

SR SR

0.147 ( 0.075 SR

NA NR

phenyl^b

phenyl^c

styryl^b

styryl^c

phenylethynyl^b

phenylethynyl^c

propyl

44% inhibition at 100 µM NR

propyl

heptyl

benzyl

60% inhibition at 100 µM NR

16.88 ( 5.19

10.69 ( 3.91

46% inhibition at 100 µM NR

3.06 ( 0.62

34.73 ( 2.7

4.63 ( 0.239

6.54 ( 0.63

11.71 ( 5.17

NR

0.545 ( 0.189 NR

30

20

52

28

42

11

NR

0.154 ( 0.041

NR

IV

II

IV

20

27.6

6.1

benzyl

0.052 ( 0.006 NR

SR NR

^aFormula, B₁, B₂, R, and n are as deﬁned in Scheme 1. Conversion % was determined by HPLC. Inhibition of platelet aggregation and selectivity

(activity at P2Y₁, P2Y₂, and P2Y₆receptors) are expressed in µM. SR ) slight response (5 to 20% of the maximal activity at highest concentration

tested, usually 100 µM); NR ) no response at highest concentration tested (usually 100 µM); ND ) not determined; NA ) not applicable. ^bMixture

of 2 diastereomers. ^cMixture of 4 diastereomers. ^dLow isolated yields relative to % conversions are a reﬂection of the degree of difﬁculty of the

puriﬁcations.

diastereomers because of the nonselective establishment of

the stereogenic acetal carbon (see Figure 2).^40,41As a result,

the dinucleoside tri and tetraphosphate monoacetals derived

from these aldehydes comprise two inseparable diastereomers,

while the diacetals comprise four inseparable diastereomers.

This pattern of selectivity was not observed for the acetals

derived from aldehydes with an sp³carbon adjacent to the

carbonyl (phenylacetaldehyde, butanal, and octanal).^42,43

For the dinucleoside polyphosphates, the acetal conﬁgu-

ration of phenylacetaldehyde diacetal 21, octanal diacetal 31,

and phenylacetaldehyde diacetal 38 was established to be

cis via 2D NMR experiments (NOESY).⁴²The data indicated

that the nature of the base (adenine in 11 vs uracil in 13)

and the number of acetals being formed had no effect on the

outcome of the acetal-forming reaction. On this basis,

assignment of the cis-acetal conﬁguration was inferred for

dinucleoside polyphosphates 20, 28–38, and 42. The con-

ﬁguration of the acetal in cytidine analogue 45 was shown

to be cis via NOE experiments.⁴²On the basis of this result,

the conﬁguration of the acetal carbon in dinucleoside

tetraphosphates 46 and 48-50 was inferred to be cis.

For compounds derived from monophosphate 1, acetal-

forming reactions provided either a single acetal diastereomer,

such as 52, or a mixture of two possible diastereomers such

as nucleotides 53–55 (Table 3). The level of diastereoselec-

tivity in these analogues could be quantiﬁed by HPLC and/

or ¹H NMR (see Experimental Details). The cis-acetal

conﬁguration of 52 was established via 2D NMR experiments

(NOESY) and was consistent with previous results in the

cytidine nucleotide series (45), the adenosine nucleotide series

(58), the dinucleoside polyphosphate series, and with the

observations of others.^40,41

As they were derived from compounds 53-55, the urea

products 60–66 were obtained as mixtures of diastereomers.

Likewise, compounds 56–59 were all the single cis isomer, as

they were derived from 52. NOE experiments were performed

on a representative of this latter group (compound 58) to

conﬁrm that the conﬁguration of the acetal carbon was

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1011

Scheme 1. General Synthesis of Dinucleoside Polyphosphate Acetals

Scheme 2. Synthesis of Diadenosine Tetraphosphate and Adenosine Analogues^a

^aReagents and conditions: (a) Phenylacetaldehyde dimethyl acetal, HCOOH, 35°C; (b) PhNCO, DMF, 45°C; (c) PhNCO, NBu₃, DMF 30°C; (d) CDI,

NBu₃, DMF, rt.

unchanged after the urea-forming reaction.⁴²Assignment of

the diastereomeric ratios for other analogues were made by

correlation and integration of selected protons in the spectra

of the mixture.

Taking advantage of the relative instability of the minor

(cis) styryl acetal diastereomer in 61 toward aqueous acid,

we cleaved its acetal to make the separation from the trans

isomer 62 possible. NOE experiments performed on 62

1012 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

Scheme 3. Synthesis of Uridine Cytidine Tetraphosphate and Cytidine Analogues^a

^aReagents and conditions: (a) Phenylacetaldehyde dimethyl acetal, HCOOH, rt; (b) p-F-PhNCO, DMF, 60°C, then 30°C; (c) p-F-PhNCO, NBu₃, DMF,

rt; (d) 5·(NBu₃)₂, DCC, DMF, 30°C; (e) phenylacetaldehyde dimethyl acetal, HCOOH, 30°C.

conﬁrmed the trans orientation of the acetal moiety in this

molecule (see Figure 2).⁴²

that led to this study was the observation that the N-methyl

anthranilic acid esters (15 and 16, Figure 1)^45,46or benzoic acid

esters (17, Figure 1) of diuridine tetraphosphate 13 modestly

antagonized ADP-induced platelet aggregation (46% inhibition

at 100 µM, IC₅₀) 9 µM, and 52% inhibition at 100 µM,

respectively). In an effort to determine whether the degree or

type of substitution was important for inhibition of platelet

aggregation activity, we decided to replace the ester moieties

with a series of acetal substituents (propyl, heptyl, phenyl,

benzyl, styryl, and ethynyl phenyl).

Results and Discussion

Platelet Aggregation Studies. While the natural dinucleoside

tetraphosphates 13 and 18 and the dinucleoside triphosphate 19

are agonists of the P2Y₂receptor and the P2Y₆receptor,⁴⁴

respectively, they are devoid of activity at any of the platelet

nucleotide receptors. Only compound 11, a potent P2Y₂agonist

(which has been found in platelet granules and is a close

analogue of the natural P2Y₁₂antagonist 3 has shown modest

potency to inhibit platelet aggregation (63% inhibition at 100

µM in a washed platelet preparation). One of the initial results

Review of the activities of the diuridine tetraphosphate

monoacetals of 13 (compounds 20, 22, 24, 26, 28, and 30)

revealed modest antagonism of platelet aggregation (Table

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1013

Scheme 4. Synthesis of Adenosine Monophosphate Analogues^a

^aReagents and conditions: (a) R₁CHO, TFA, or HCOOH, rt; (b) PhNCO, DMF, 45°C; (c) R₂NCO, DMF.

1). These diuridine tetraphosphate monoacetals demonstrated

higher potency for inhibiting ADP-induced platelet aggrega-

tion than monoester 15. There appeared to be a slight

preference for the styryl (24) and ethynyl phenyl (26)

derivatives. Interestingly, benzyl acetal diuridine triphosphate

35 (34.73 µM) was found to be comparable in potency to its

tetraphosphate 20, suggesting that there is some tolerance

for variation in the polyphosphate chain length. Consistent

with the trend observed for esters 15 and 16, substitution of

diuridine tetraphosphate 13 with two lipophilic acetals

(compounds 21, 23, 25, 26, 29, and 31, Table 1) provided

analogues with slightly improved potency when compared

with their monosubstituted counterparts. Notably, the slight

preference for the styryl and ethynyl phenyl derivatives

remained consistent. The dibenzyl acetal diuridine triphos-

phate 36 also showed good potency as an inhibitor of ADP-

induced platelet aggregation (IC₅₀) 4.63 µM), which was

consistent with that observed for the analogous tetraphosphate

21 (IC₅₀) 1.82 µM).

Modiﬁcation of diadenosine tetraphosphate 11 with either

one or two benzyl acetals (37 and 38) provided potencies

comparable to those of analogues derived from diuridine

tetraphosphate13.Lastly,fortheunsymmetricalcytidine-uridine

dinucleoside tetraphosphate 18, the two possible monobenzyl

acetals (32 and 33) were only weakly active in the platelet

aggregation assay, while the corresponding dibenzyl acetal

34 had a potency of 3.06 µM. This value was comparable to

that of the most potent analogues in the diuridine tetraphos-

phate acetal series. These results established the importance

of polysubstitution on the 2′,3′-ribose hydroxyl groups for

the enhancement of potency and conﬁrmed that more

chemically stable acetal substituents could replace the

N-methyl anthranilic ester moieties without the loss of

potency.⁴⁷

ﬁcations to diadenosine tetraphosphate 11 and uridine-cytidine

tetraphosphate 18 provided diacetals of comparable potency

to those derived from 13, and these molecules offered an

exocyclic amino group as an additional point of modiﬁcation.

Accordingly, modiﬁcation of 38 to a phenyl urea at the

6-amino position provided diurea 42. Remarkably, 42 was

20-fold more potent than its unsubstituted diadenosine

tetraphosphate diacetal core 38 (0.52 vs 11.71 µM). Similarly,

conversion of 34 to a p-ﬂuoro-phenylurea in the 4-position

resulted in 50, which was 60-fold more potent than its

cytidine-uridine tetraphosphate diacetal core 34 (0.048 vs

3.06 µM). This dramatic effect led us to further examine the

contributions of the benzyl acetal and the p-ﬂuoro-phenylurea

moieties on the potency of the cytidine-uridine tetraphos-

phate core. The p-ﬂuoro-phenylurea of Up₄C (47) was found

to be inactive. The monobenzyl acetal (U)/p-ﬂuoro-pheny-

lurea of Up₄C (48) was found to have an IC₅₀) 23.93 µM,

while the monobenzyl acetal (C)/p-ﬂuoro-phenylurea of Up₄C

(49) was found to have an IC₅₀) 12.92 µM. Interestingly,

it was essential for both riboses to be converted to acetals in

the dinucleoside tetraphosphate core to achieve high nano-

molar activity.

We next examined this dramatic substituent effect on CMP

(43). Acylation of 43 with p-ﬂuorophenylisocyanate or conver-

sion of 43 to the corresponding cis-acetal with phenylacetal-

dehyde produced 44 and 45, respectively, which were inactive.

However, the combination of both modiﬁcations converted 43

into a micromolar antagonist (46, IC₅₀) 11.17 µM, Table 2).

Similar modiﬁcations of ADP (2), a natural agonist of P2Y₁₂

receptor, gave a comparable SAR pattern, albeit with a more

favorable outcome. The cis-benzyl acetal (40) was found to be

a modestly active inhibitor. Likewise, modiﬁcation of 2 with a

phenylurea moiety (39) also converted the agonist into an

antagonist. The incorporation of both of these lipophilic

modiﬁcations led to 41, with potency for inhibition of

aggregation of 0.052 µM, comparable to the best of the

dinucleoside tetraphosphate series (50). These data are

summarized in Table 2.

None of the tested lipophilic modiﬁcations at the ribose

hydroxyl groups of 13 led to compounds with sufﬁcient

potency as platelet aggregation inhibitors to be advanced

further into development. However, the benzyl acetal modi-

1014 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

Table 2. Inhibition of Platelet Aggregation and Selectivity for Dinucleoside Tetraphosphate Acetal and Urea Analogues, and Their Synthetic

Intermediates^a

^aPlatelet aggregation, inhibition, and selectivity (agonist activity at P2Y₁, P2Y₂, and P2Y₆receptors) are expressed in µM. SR ) slight response (5 to

20% of the maximal activity at highest concentration tested, usually 100 µM); NR ) no response at highest concentration tested (usually 100 µM); NT )

not tested; NA ) not applicable.

On the basis of the results for 41 and 46, we chose to examine

similar modiﬁcations for 1 (Scheme 4). Only those acetal

substitutions that produced the more potent compounds of the

dinucleoside polyphosphate series were prepared and tested for

potency. By examining the potency of substituted adenosine

diphosphate 41 and cytidine monophosphate 46, it was clear

that the purine core is an essential component for maintaining

the potency for the inhibition of aggregation. Furthermore,

although highly potent antagonists would likely have been found

in series based on 2 (or even 3), the chemical instability of the

di and triphosphate groups was deemed a signiﬁcant drug

development obstacle.⁴⁸

Much as was seen for phenylurea analogue 39, the reaction

of 1 with phenylisocyanate converted an inactive compound into

a modestly active antagonist (51). Similarly, the formation of

acetals of 1 (phenyl, benzyl, styryl, and ethynyl phenyl) also

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1015

Table 3. Inhibition of Platelet Aggregation and Selectivity for Adenosine Acetal and Urea Analogues^a

^aR₁and R₂are as deﬁned in Scheme 4. Inhibition of platelet aggregation is expressed in µM. SR ) slight response (5 to 20% of the maximal activity

at highest concentration tested, usually 100 µM); NR ) no response at highest concentration tested (usually 100 µM); NA ) not applicable.

provided analogues that were modestly active (52–55). Prepara-

tion of the monophosphate analogue of 41 (phenylurea and

benzyl acetal, 56) showed an approximately 3-fold loss in

potency (0.052 vs 0.160 µM, respectively), which suggested

that truncating the phosphate chain would not lead to an

insurmountable loss of potency. Comparison of the phenyl urea

moiety in the acetal series benzyl, styryl, and ethynyl phenyl

(56, 64, and 66, respectively) did not reveal any obvious

advantage of any of these acetals for increased potency.

A switch to aliphatic ureas led to a breakthrough in potency

for the monophosphate series. However, inconsistent trends

were found when comparisons were drawn across different

acetal series. The ethyl urea derivatives 58 (cis-benzyl acetal),

60 (mixture of phenyl acetals), 61 (mixture of styryl acetals),

and 62 (trans-styryl acetal) were 0–5-fold more potent than

diphosphate 41, making them the most potent compounds in

this study (0.041, 0.013, 0.011, and 0.016 µM, respectively).

Preparation of three n-hexyl urea derivatives 57 (cis-benzyl

1016 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

On the basis of these results, the described modiﬁcations of

the nucleotides and dinucleoside polyphosphates described

herein confer antagonist properties to these analogues and

eliminate the P2Y₂and P2Y₁activity, achieving selectivity for

the inhibition of platelet aggregation toward the P2Y₁₂receptor.

Dinucleoside polyphosphates and nucleotides modiﬁed with

acetals and ureas were able to inhibit the aggregation of platelets

in a concentration-dependent manner. The inhibitory effects of

these compounds have been demonstrated using several platelet

preparations, including whole blood, platelet rich plasma (PRP),

and washed platelet preparations. The potency of a given

compound in these assays varies among the different platelet

preparations, mainly because of differences in the protein-

binding characteristics of the compound. In general, the apparent

potency observed for the compounds of this study was highest

in the washed platelet preparation followed by whole blood and

PRP. For example, compound 62 exhibited a potency of 0.016

µM in the washed platelet assay, while the potencies in whole

blood and PRP were 0.180 µM and 0.760 µM, respectively.

All data shown in this study were obtained with the washed

platelet preparation.

Figure 2. Acetal diastereoisomerism.

acetal), 63 (mixture of styryl acetals), and 65 (mixture of

ethynyl phenyl acetals) led to derivatives with a clear

preference for the benzyl acetal (0.053, 0.237, and 0.211 µM,

respectively). Comparison of three alkyl ureas in the benzyl

acetal series n-hexyl (57), cyclopentyl (59), and ethyl (58)

revealed no preference for any of the urea alkyl groups

(0.053, 0.036, and 0.041 µM, respectively). On the basis of

pharmacological, physicochemical, and synthetic consider-

ations, compound 62 was nominated as a candidate for further

development as an inhibitor of platelet aggregation. A full

pharmacological and biochemical characterization of this

compound will be reported elsewhere.

In order to study the nature of the inhibition of platelet

aggregation, we used the diuridine tetraphosphate di-2′,3′-

phenylacetaldehyde acetal derivative 21. Compound 21 pro-

duced a dose-dependent shift to the right in the platelet

aggregation concentration–response curve of the ADP analogue

2-MeS-ADP (Figure 3A). The Schild analysis⁵⁰suggested that

the antagonism demonstrated by 21 on platelet aggregation is

competitive (Figure 3B), with a K_Bof 0.249 µM and a slope of

0.742. The reversibility of the effects of 21 on platelet

aggregation was studied in a preparation of washed platelets.

Platelets were treated with 50 µM 21 for 3 min at 37 °C in the

absence of ﬁbrinogen, then were either stimulated with 2-MeS-

ADP and ﬁbrinogen or were washed once by centrifugation

before stimulation with 2-MeS-ADP and ﬁbrinogen (Figure 4A

and 4B). Unlike the thienopyridines 9 and 10, the compounds

reported here, as exempliﬁed by compound 21, behaved as

competitive and reversible antagonists of platelet aggregation

mediated by the P2Y₁₂receptor.

P2Y Receptor Selectivity. Since inhibition of ADP-induced

platelet aggregation could result from the blockade of either

the P2Y₁or the P2Y₁₂receptors, we ruled out any activity of

these compounds at the P2Y₁receptor by conducting calcium

mobilization assays in 1321N1 human astrocytoma cells ex-

pressing the recombinant human P2Y₁receptor (Tables 1, 2,

and 3). None of the compounds that inhibited ADP-induced

platelet aggregation had any effect as an antagonist at the P2Y₁

receptor cell line, suggesting that the inhibition of aggregation

by ADP was mediated by the antagonism of the P2Y₁₂receptor.

This observation was consistent with the lack of effect of these

compounds on the P2Y₁receptor-mediated platelet shape change

induced by ADP and further conﬁrmed in a selected number of

compounds by the inhibition of the P2Y₁₂receptor response in

C6 rat glioma cells, which express the native P2Y₁₂receptor.⁴⁹

The diuridine tetraphosphate monoacetals 20, 22, 24, 26, 28,

and 30 were all P2Y₂agonists, exhibiting potencies up to 6-fold

lower than that of 13. Consistent with this trend, the diadenosine

tetraphosphate monobenzyl acetal 37 was slightly less potent

as a P2Y₂agonist than 11. Similarly, the cytidine-uridine

tetraphosphate monoacetal (C), 32, was approximately 10-fold

less potent at P2Y₂than 18; however, the cytidine-uridine

tetraphosphate monoacetal (U), 33, was inactive. Dinucleoside

triphosphate monoacetal 35 showed slightly higher agonist

potency at P2Y₆than 19 (0.154 vs 0.40 µM, Table 1). The

corresponding diacetal 36 was not active at this receptor.

Likewise, diacetal analogues derived from the three dinucleoside

tetraphosphate cores showed a marked reduction in agonist

potency toward the P2Y₂receptor (Table 1). For example, 27

exhibited a 10-fold loss of agonist potency (EC₅₀(P2Y₂) )

0.350 µM) in a calcium mobilization assay, relative to core 13.

The other diacetals in this series were even less active at P2Y₂

than 27 (compounds 21 and 23) or showed little to no activity

(compounds 25 and 29). The dibenzyl acetals 34 and 38 were

also inactive at P2Y₂. Taken together, these results suggest that

increasing lipophilic modiﬁcation of the hydroxyl groups of

dinucleoside polyphosphates leads to both increased potency

and selectivity toward the platelet P2Y₁₂receptor.

Conclusions

These studies showed that acetal substitution on the 2′,3′-

ribose moiety of dinucleoside polyphosphates and nucleotides

was an important determinant in conferring antagonist properties

toward the P2Y₁₂receptor. Indeed, this effect was robust enough

that dinucleoside polyphosphates such as 13 and 18, which

normally do not interact with the P2Y₁₂receptor as either an

agonist or antagonist, were converted via these modiﬁcations

into strong antagonists. Expansion of this concept around

dinucleoside polyphosphate 13 demonstrated that a variety of

substituents on the acetal moiety were well tolerated, giving

structurally diverse analogues on which a SAR campaign could

be based. We found that the number of acetals (lipophilic

substitutions) correlated directly with the potency of the

molecule to inhibit aggregation, conﬁrming the observations

made with initial screening hits, 15 and 16. Furthermore, there

was a direct relationship between enhanced P2Y₁₂antagonist

potency with increasing lipophilic substitution and decrease in

agonist potency at several other P2Y receptors, suggesting that

both potent and selective compounds might be developed in

this series.

In the nucleotide series, monophosphate 56 showed greater

selectivity over the P2Y₁, P2Y₂, and P2Y₆receptors versus

its diphosphate counterpart 41. In addition, it was not active

at the P2Y₄receptor. This trend held true for compounds

51–66 (Table 3).

We further identiﬁed the exocyclic amino group of adenine

and cytidine-containing dinucleoside polyphosphates and nucle-

otides as a second site for lipophilic modiﬁcation, where ureas

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1017

Figure 3. Effect of compound 21 on 2-MeS-ADP-promoted platelet aggregation. (A) Platelet aggregation was measured as described in General

Methods. Assays were in the presence of the indicated concentrations of 2-MeS-ADP alone or in the presence of 1, 3, 10, 30, or 100 µM compound

21. (B) Arunlakshana-Schild plot from the data presented in A. The slope of the plot is 0.7425, and K_Bis 0.251 µM.

Figure 4. Reversible inhibition of platelet aggregation by compound 21. Platelets were incubated in the absence (control) or in the presence of 50

µM of 21. In panel A, after a three minute incubation with drug, the indicated concentrations of 2-MeS-ADP and ﬁbrinogen were added, and

platelet aggregation was monitored during the following 6–8 min. In panel B, after the three minute incubation with drug, both control and compound

21-treated platelets were centrifuged, the supernatant was removed, and the platelets were resuspended in drug-free media. Following this, the

indicated concentrations of 2-MeS-ADP and ﬁbrinogen were added, and the aggregation was monitored as described above.

When resolution was still unsatisfactory, additional experiments

were carried out in CD₃OD or DMSO-d₆.

in this position greatly enhanced the antagonist potency. In most

cases, modiﬁcation of either a ribose with a 2′,3′-acetal moiety

or a base with a urea was enough to generate at least some

antagonist behavior. However, these modiﬁcations proved to

be synergistic, and the disubstituted compounds were the more

potent antagonists of platelet aggregation. The compounds

reported here are competitive and reversible antagonists of the

P2Y₁₂receptor. While some of the dinucleoside polyphosphates

were potent antagonists of aggregation (compounds 42 and 50),

we found that the substituted nucleotide 41, used to prepare

42, was more potent than its dinucleoside polyphosphate

product. Optimization of 41 led to a series derived from

adenosine 5′-monophosphate (1), of which one, 62, was selected

for in vivo preclinical studies. On the basis of positive results

from animal studies, compound 62 was nominated for further

development, where it was subsequently advanced to clinical

trials.⁵¹This study provides the basis for further development

of potent and selective antagonists of the P2Y₁₂receptor that

should prove useful for the treatment of cardiovascular diseases.

Analytical HPLC chromatograms were obtained on either Waters

600S or Waters Alliance 2795 equipped with diode array detectors.

The C₁₈columns used (either Waters Xterra RP₁₈, 150 × 4.6 mm,

3.5 µm, or Alltech Apollo C₁₈) were run employing a linear gradient

from 0.02 M KH₂PO₄/0.005 M NBu₄⁺HSO₄(pH adjusted to 5.0

with KOH) to 90% ACN 10% water over 20 min (1 mL/min), a

linear gradient from 0.1% TFA to ACN over 20 min (1 mL/min),

or a linear gradient from 0.05 M NH₄OAc (pH 6) to ACN over 20

min (1 mL/min). The former ion pairing system allowed the analysis

of unmodiﬁed parent dinucleoside polyphosphates to be carried out

under reverse-phase conditions. It also was employed to detect the

presence or absence of diastereomers arising from the acetal

moieties and to quantify the diasteromeric ratios in compounds

53–55 and 60–66. The NH₄OAc and TFA systems generally did

not provide information on diastereomeric ratios but were used to

assess the overall purity of the modiﬁed nucleotides. Purities of

test compounds exceeded 95% and in most cases 98%, as measured

in these systems.

Unmodiﬁed dinucleoside polyphosphates 11, 13, 18, and 19 were

puriﬁed using either Waters Delta 600 or Prep LC 2000 with diode

array detection, in conjunction with an ion-exchange preparative

column (Hamilton PRP X-100 anion exchange column, 20 µm;

gradient from water to 1 M ammonium bicarbonate, 100 mL/min).

Alternately, these molecules were puriﬁed using DEAE Sephadex

in a medium pressure mode, utilizing a gradient from water to 1

M ammonium bicarbonate. Dinucleoside polyphosphates and nucle-

otides modiﬁed with lipophilic substituents were generally puriﬁed

by preparative RPHPLC (Waters PrepLC system, equipped with

either two or three 25 × 100 mm or two 40 × 100 mm Waters

NovaPak C₁₈segments, depending on the scale of the reaction to

Experimental Details

General Methods. All reagents were acquired from commercial

suppliers and were generally used without further puriﬁcation. All

reactions requiring anhydrous conditions were performed under

nitrogen, using anhydrous solvents packaged in septum bottles. ¹H

and ³¹P NMR spectra were obtained at 300 and 121.5 MHz

respectively, using Varian Gemini 2000. Generally, samples run

in D₂O were pretreated with a chelating resin (Chelex 100 Na) in

water, followed by lyophilization. This was done to remove

paramagnetic ions in an effort to improve the NMR resolution.

1018 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

be puriﬁed; gradient from 0.05 M NH₄OAc to ACN, 20–70 mL/

min, diode array detection) and isolated as the ammonium salts.

Alternately, products from small-scale reactions were separated on

a semipreparative reverse-phase column (Alltech Nucleotide/

Nucleoside C₁₈, 10–250 mm, 7 µm, gradient from 0.1 M NH₄OAc

to MeOH, 5 mL/min, diode array detector). Irrespective of the

preparative RPHPLC puriﬁcation method, dry products were

obtained by partial evaporation, followed by lyophilization.

LR-LCMS were obtained using a Waters Micromass ZQ mass

spectrograph (ESI+ and ESI-) coupled to a Waters Alliance 2695

HPLC. For the lipophilic dinucleoside polyphosphate and nucle-

otides 15–42 and 44–66, a linear gradient from 0.05 M NH₄OAc

(pH 6) to ACN over 20 min (0.5 mL/min) was used.

Platelet Aggregation Assay. Blood was collected from healthy

volunteers into syringes containing 1/6 ﬁnal blood volume of

anticoagulant ACD (65 mM citric acid, 85 mM sodium citrate, and

110 mM dextrose). The blood was centrifuged at 180g for 15 min,

and the supernatant (platelet rich plasma) was removed. The platelet

rich plasma was centrifuged and the platelet pellet resuspended in

a buffer consisting of (mM) NaCl (137), KCl (2.7), CaCl₂(2) MgCl₂

(1), NaH₂PO₄(3), glucose (5), HEPES (10) at pH 7.4, and 0.2%

bovine serum albumin. These centrifugations and washes were

repeated twice followed by resuspension in the media described

above containing 0.25 U apyrase/mL. To prevent platelet activation

during the procedure, all platelet resuspensions were conducted in

the presence of 5 µM PGI₂, followed by incubation at 37 °C for 10

min. Platelet suspensions were further supplemented with the same

concentration of PGI₂before each centrifugation. Platelet aggrega-

tion was measured using the optical mode of a ChronoLog

aggregometer (Havertown, PA). Five hundred microliters of platelet

suspension containing 1 mg/mL ﬁbrinogen was warmed to 37 °C

and stirred at 1000 rpm. Several concentrations of test compound

(usually between 1 nM and 100 µM) were added to the platelet

suspension and incubated for two minutes, followed by the addition

of a maximally effective concentration of ADP (EC₉₀; usually

between 1 and 10 µM ADP). The aggregation was recorded for 8

min.

Calcium Mobilization Assay. For calcium mobilization assays,

1321N1 human astrocytoma cells were loaded with a solution of

Fluo-3 AM (2.5 µM ﬁnal concentration) in an assay buffer

consisting of (mM) KCl (10), NaCl (118), CaCl₂(2.5), MgCl₂(1),

HEPES (20), and glucose (10) at pH 7.4. After a 60-min incubation

with Fluo-3 AM at 25 °C, cells were washed and stimulated with

the indicated concentrations of P2Y receptor agonists. Intracellular

calcium levels were simultaneously monitored in each well by

measuring the changes in ﬂuorescence intensity using FLIPR

(Molecular Devices Corp., Sunnyvale, CA).

Data Analysis and Statistics. Data for selected compounds were

expressed as the mean ( standard error of the mean (SEM). The

potency of inhibitors of platelet aggregation was calculated from

the inhibition of ADP-induced aggregation obtained at each

concentration of test compound by ﬁtting the data to a four-

parameter logistic equation using the GraphPad software package

(GraphPad Corp. San Diego, CA). When a broad range of

concentrations of P2Y₁₂antagonist were tested (usually from 1 nM

to 100 µM), an IC₅₀value was also obtained. IC₅₀values represent

the concentration of antagonist needed to inhibit by 50% the

aggregation elicited by a given concentration of ADP.

The nucleotides purchased as their sodium salts were ﬁrst

converted to their free acid forms by treatment with ﬁve weight-

equivalents of Dowex 50 H⁺in water. After stirring for 10 min,

the strongly acidic solutions were ﬁltered and neutralized with an

excess of NBu₃in MeOH. Once a neutral pH or greater was

obtained, the solvents were evaporated to dryness and the residues

were coevaporated with dry DMF. The solids were dried to constant

weight on a lyophilizer.

Tributylamine content in the solids was determined by ¹H NMR,

and the molecular weights were adjusted accordingly. When fully

dried, nucleoside monophosphates generally had 1 mol equivalent

of NBu₃, 1.33–1.5 mol equivalents of diphosphates, and 2 mol

equivalents of triphosphates. When easily handled solids were

obtained, they were added as such to the dinucleoside polyphos-

phate-forming reactions described below; oily tributylammonium

salts were dissolved to known concentrations in DMF.

The dinucleoside polyphosphates used as starting materials (11,

13, 18, and 19) were synthesized from the appropriate nucleotide

subunits, according to the general methods described in a previous

communication.⁴⁴Brieﬂy, for the synthesis of 11, nucleotide 3 in

the ditributylammonium salt form was ﬁrst activated with dicyclo-

hexylcarbodiimide in DMF at room temperature, which converted

the linear nucleoside triphosphate to the known cyclical trimeta-

phosphate.³⁹Following this, 1 (as monotributylammonium salt) was

added, and the reaction mixture was heated at 35 °C for 72 h. The

dinucleoside polyphosphate so obtained (11) was isolated via ion

exchange prep HPLC. Similarly, dinucleoside polyphosphates 13

and 18 were prepared via the DCC activation of 6 (2NBu₃salt),

followed by the reaction of the activated species with the appropriate

nucleoside monophosphate in the NBu₃salt form (4 for dinucleoside

polyphosphate 13, and 43 for 18). Dinucleoside triphosphate 19

was prepared from the reaction between commercially available

uridine-5′-monophosphomorpholidate-4-morpholine-N,N′-dicyclo-

hexylcarboxamidine salt and the NBu₃salt of 5 in DMF. The typical

yields for these procedures ranged from 10 to 40%. Two dinucleo-

side polyphosphates (11 and 47) were converted to their free acid

forms as a prelude to further synthesis. This was accomplished by

treating the NBu₃salt forms with Dowex 50 H⁺as described above

for mononucleotide sodium salts, followed by direct freezing and

lyophilization of the resultant dinucleoside free acids. All free acid

forms were stored at –20 °C to prevent decomposition.

Synthesis. P¹-[(2′(3′)-O-2-(Methylamino)benzoyl)uridine-5′-

yl]-P⁴-[uridine-5′-yl]-tetraphosphate (15) and P¹-[(2′(3′)-O-2-

(Methylamino)benzoyl)uridine-5′-yl]-P⁴-[(2′(3′)-O-2-(methylami-

no)benzoyl)uridine-5′-yl]-tetraphosphate (16). Up4U and 4Na salt

(13, 1.0 g, 1.14 mmol) was dissolved in water (3 mL) and DMF (3

mL). To this was added N-methyl isatoicanhydride (303 mg, 1.71

mmol) and the reaction mixture heated at 50 °C for 5 h. The reaction

mixture was diluted to 15 mL with water, and the insoluble material

was removed by ﬁltration. The products were separated via

preparative HPLC. The yield of the mono MANT analogue 15 was

350 mg (31%), while the yield of the di MANT analogue 16 was

1

430 mg (33.5%). 15: H NMR (300 MHz, D₂O) δ 2.70 (d, 3H),

4.12 (m, 7H), 4.39 (s, 1H), 4.53 (t, 1H), 5.34 (m, 1H), 5.72 (m,

2H), 5.83 (dd, 1H), 6.01 (m, 1H), 6.53 (q, 1H), 6.67 (t, 1H), 7.35

(t, 1H), 7.78 (m, 3H). ³¹P NMR (121.47 MHz, D₂O) δ -21.84(m,

2P), -10.20 (m, 2P). MS (ES): m/z 921.9 (M - H). 16: ¹H NMR

(300 MHz, D₂O) δ 2.66 (t, 6H), 4.22 (m, 6H), 4.46 (q, 2H), 5.22

(t, 1H), 5.28 (d, 1H), 5.72 (d, 8.1 Hz, 1H), 5.84 (d, 1.5H), 5.93 (m,

1.5H), 6.54 (m, 4H), 7.28 (m, 2H), 7.77 (m, 4H). ³¹P NMR (121.47

MHz, D₂O) δ -21.76 (m, 2P), -10.16 (m, 2P). MS (ES): m/z

1055.1 (M - H).

General Method for the Preparation of Tributylammonium

Salts of Dinucleoside Polyphosphates and Nucleotides. The

nucleotides used for the procedures described herein were obtained

from commercial sources as the recrystallized free acids [adenosine

5′-monophosphate (1) and cytidine 5′-monophosphate (43)] or as

the sodium salts [uridine 5′-monophosphate (4), adenosine 5′-

diphosphate (2), uridine 5′-triphosphate (6), and adenosine 5′-

triphosphate (3)]. As a prelude to the dinucleoside polyphosphate

syntheses, the nucleotide free acids were converted to their

tributylammonium salts by treatment with an excess of NBu₃in

aqueous methanol. After stirring for 20 min, the solutions were

evaporated to dryness and the residue coevaporated with dry DMF.

Solids were obtained with overnight drying on a lyophilizer.

P¹-[(2′,3′)-di-O-Benzoyl)uridine-5′-yl]-P⁴-[(2′,3′)-di-O-benzo-

yl)uridine-5′-yl]-tetraphosphate (17). Compound 13, ditributylam-

monium salt (200 mg, 0.172 mmol), and 4-dimethylaminopyridine

(5 mg, 0.04 mmol) were dissolved in pyridine (3.0 mL) and benzoic

anhydride (500 mg, 2.21 mmol) added. The reaction mixture was

stirred overnight at room temperature, after which the pyridine was

removed. The residue was partitioned between ethyl acetate (75

mL) and 0.5 M NaHCO₃(50 mL). The layers were separated and

the aqueous layer containing the products concentrated to 8 mL.

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1019

The products were separated via preparative HPLC. The isolated

yield of the tetra-benzoate (17) was 64 mg (31%). Only minimal

quantities of the mono and dibenzoates and a lesser amount of the

tribenzoate were produced under the chosen reaction conditions

(m, 2H), 6.78 (m, 2H), 7.26 (m, 6H), 7.38 (m, 4H), 7.68 (m, 2H).

³¹P NMR (121.47 MHz, D₂O) δ -22.07 (m, 2P), -10.72 (m, 2P).

MS (ES): m/z 1017.3 (M - H).

P¹-[(2′,3′)-O-(Phenylethynylmethylidene)uridine-5′-yl]-P⁴-

[uridine-5′-yl]-tetraphosphate (26) and P¹-[(2′,3′)-O-(Phenyl-

ethynylmethylidene)uridine-5′-yl]-P⁴-[(2′,3′)-O-(phenylethynyl-

methylidene)uridine-5′-yl]-tetraphosphate (27). Compound 13

and 4Na salt (250 mg, 0.28 mmol) were dissolved in 99% formic

acid (2 mL) and phenylpropargyl aldehyde (105 µL, 3.42 mmol)

added. The reaction was stirred for 1 h, and the formic acid was

removed by evaporation. The residue was partitioned between

EtOAc and 1 M NaHCO₃, and the products in the aqueous layer

puriﬁed via preparative HPLC. The isolated yield of the monoacetal

(26) was 82 mg (32%) and of the diacetal (27) 10 mg (3.5%). 26:

¹H NMR (300 MHz, D₂O) δ 4.13 (m, 5H), 4.24 (m, 2H), 4.49 (s,

0.5H), 5.03 (d, 2H), 5.82 (m, 3.5H), 6.00 (s, 0.5H), 6.13 (s, 1H),

7.31 (m, 3H), 7.44 (m, 2H), 7.76 (m, 2H). ³¹P NMR (121.47 MHz,

D₂O) δ -22.13 (m, 2P), -10.49 (m, 2P). MS (ES): m/z 901.2 (M

1

and were not isolated. 17: H NMR (300 MHz, D₂O) δ 4.30 (m,

4H), 5.63 (t, 2H), 5.72 (m, 2H), 5.90 (d, J ) 8.1 Hz, 2H), 6.21 (d,

J ) 5.4 Hz, 2H), 7.27 (m, 8H), 7.41 m, 4H), 7.59 (d, 4H), 7.71 (d,

4H), 7.93 (d, J ) 8.1 Hz, 2H). ³¹P NMR (121.47 MHz, D₂O) δ

-21.95 (m, 2P), -10.50 (m, 2P). MS (ES): m/z 1205.3 (M - H).

P¹-[(2′,3′)-O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[uridine-5′-

yl]-tetraphosphate (20) and P¹-[(2′,3′)-O-(2-Phenylethylidene)-

uridine-5′-yl]-P⁴-[(2′,3′)-O-(2-phenylethylidene)uridine-5′-yl]-

tetraphosphate (21). Compound 13 and 4Na salt (290 mg, 0.332

mmol) were dissolved in 99% formic acid (3 mL) and phenylac-

etaldehyde dimethyl acetal (110 µL, 0.664 mmol) added. The

reaction was stirred for 48 h and the formic acid removed by

evaporation. The residue was partitioned between butyl acetate (15

mL) and 0.7 M NaHCO₃(15 mL). The layers were separated and

the products in the aqueous layer puriﬁed via preparative HPLC.

The isolated yield of the monoacetal (20) was 88 mg (30%), and

1

- H). 27: H NMR (300 MHz, D₂O) δ 4.15 (m, 4H), 4.42 (m,

2H), 5.04 (m, 4H), 5.70 (t, 1H), 5.78 (m, 2H), 5.95 (s, 0.5H), 5.98

(s, 0.5H), 6.01 (s, 0.5H), 6.08 (s, 0.8H), 6.13 (d, 0.6H), 7.29 (m,

5H), 7.40 (m, 5H), 7.67 (m, 2H). ³¹P NMR (121.47 MHz, D₂O) δ

-22.07 (m, 2P), -10.72 (m, 2P). MS (ES): m/z 1013.2 (M - H).

P¹-[(2′,3′)-O-(Propylmethylidene)uridine-5′-yl]-P⁴-[uridine-5′-

yl]-tetraphosphate (28) and P¹-[(2′,3′)-O-(Propylmethylidene)-

uridine-5′-yl]-P⁴-[(2′,3′)-O-(propylmethylidene)uridine-5′-yl]-

tetraphosphate (29). Compound 13 and 4Na salt (1.0 g, 1.14 mmol)

was dissolved in 99% formic acid (4 mL) and butyraldehyde (304

µL, 3.42 mmol) added. The reaction was stirred for 16 h, and the

formic acid was removed by evaporation. The residue was

partitioned between EtOAc and 1 M NaHCO₃, and the products in

the aqueous layer puriﬁed via preparative HPLC. The yield of the

monoacetal (28) was 227 mg (23%) and of the diacetal (29) 309

mg (30%). 28: ¹H NMR (300 MHz, D₂O) δ 0.80 (t, 3H), 1.28 (m,

2H), 1.61 (m, 2H), 4.08 (m, 5H), 4.22 (m, 2H), 4.48 (s, 1H), 4.79

1

of the diacetal (21) 60 mg (18%). 20: H NMR (300 MHz, D₂O)

δ 2.99 (d, 2H), 4.04 (m, 5H), 4.22 (m, 2H), 4.32 (s, 1H), 4.78 (m,

2H),5.30 (t, 1H), 5.36 (d, J ) 2.4 Hz, 1H), 5.73 (d, J ) 8.1 Hz,

1H), 5.81 (t, 2H), 7.20 (m, 5H), 7.63 (d, J ) 8.1 Hz, 1H), 7.79 (d,

J ) 8.1 Hz, 1H). ³¹P NMR (121.47 MHz, D₂O) δ -22.00 (m, 2P),

-10.78 (m, 1P), -10.18 (m, 1P). MS (ES): m/z 890.9 (M - H).

21: ¹H NMR (300 MHz, D₂O) δ 2.98 (d, J ) 3.9 Hz, 4H), 3.98 (s,

4H), 4.27 (s, 2H), 4.74 (m, 4H), 5.28 (t, J ) 3.9 Hz, 2H), 5.37 (d,

J ) 2.7 Hz, 2H), 5.74 (d, J ) 8.1 Hz, 2H), 7.21 (m, 10H), 7.61 (d,

J ) 8.1 Hz, 2H).4 ³¹P NMR (121.47 MHz, D₂O) δ -21.81 (m,

2P), -10.57 (m, 2P). MS (ES): m/z 993.0 (M - H).

P¹-[(2′,3′)-O-(Phenylmethylidene)uridine-5′-yl]-P⁴-[uridine-5′-

yl]-tetraphosphate (22) and P¹-[(2′,3′)-O-(Phenylmethylidene)-

uridine-5′-yl]-P⁴-[(2′,3′)-O-(phenylmethylidene)uridine-5′-yl]-

tetraphosphate (23). Compound 13 and 4Na salt (500 mg, 0.569

mmol) were dissolved in 99% formic acid (4 mL) and benzaldehyde

dimethyl acetal (342 µL, 2.27 mmol) added. The reaction was stirred

for 4 h, then the formic acid removed by evaporation. The residue

was partitioned between ethyl acetate and 1 M NaHCO₃and the

products in the aqueous layer puriﬁed via preparative HPLC. The

yield of the monoacetal (22, 48839) was 211 mg (42%), and of

the diacetal (23, 48840) 37 mg (6.7%). 22: ¹H NMR (300 MHz, D₂O)

δ 4.07 (m, 7H), 4.43 (m, 0.4H), 4.60 (m, 0.6H), 4.97 (d, 2H), 5.73

(m, 3H), 5.86 (d, 1.6H), 6.00 (s, 0.4H), 7.36 (m, 5H), 7.64 (d, 0.4H),

7.72 (t, 1.6H). ³¹P NMR (121.47 MHz, D₂O) δ -22.01 (m, 2P),

-10.77 (m, 0.66P), -10.32 (m, 1.33P). MS (ES): m/z 877.2 (M -

H). 23: ¹H NMR (300 MHz, D₂O) δ 4.06 (m, 4H), 4.36 (s, 0.7H),

4.54 (s, 1.3H),4.91 (m, 4H), 5.69 (t, 2H), 5.80 (d, 2H), 5.85 (s,

1.3H), 5.93 (s, 0.7H), 7.31 (m, 10H), 7.57 (d, 0.7H), 7.67 (d, 1.3H).

³¹P NMR (121.47 MHz, D₂O) δ -21.89 (m, 2P), -10.69 (m, 2P).

MS (ES): m/z 965.1 (M - H).

(m, 2H), 5.06 (t, 1H), 5.77 (m, 4H), 7.70 (d, 1H), 7.80 (d, 1H). ³¹

P

NMR (121.47 MHz, D₂O) δ -21.95 (m, 2P), -10.68 (m, 1P),

-10.23 (m 1P). MS (ES): m/z 843.3 (M - H). 29: ¹H NMR (300

MHz, D₂O) δ 0.80 (t, 6H), 1.29 (m, 4H), 1.62 (m, 4H), 4.06 (m,

4H), 4.47 (s, 2H), 4.79 (m, 4H), 5.06 (t, 2H), 5.76 (m, 4H), 7.70

(d, 2H). ³¹P NMR (121.47 MHz, D₂O) δ -21.92 (m, 2P), -10.63

(m, 2P). MS (ES): m/z 897.2 (M - H).

P¹-[(2′,3′)-O-(Heptylmethylidene)uridine-5′-yl]-P⁴-[uridine-5′-

yl]-tetraphosphate (30) and P¹-[(2′,3′)-O-(Heptylmethylidene)-

uridine-5′-yl]-P⁴-[(2′,3′)-O-(heptylmethylidene)uridine-5′-yl]-

tetraphosphate (31). Compound 13 and 4Na salt (1.0 g, 1.14 mmol)

were dissolved in 99% formic acid (4 mL) and butyraldehyde (533

µL, 3.42 mmol) added. The reaction was stirred for 2.25 h, then

the formic acid was removed by evaporation. The residue was

partitioned between EtOAc and 1 M NaHCO₃and the products in

the aqueous layer puriﬁed via preparative HPLC. The yield of the

monoacetal (30) was 253 mg (25%) and of the diacetal (31) 375

mg (32%). 30: ¹H NMR (300 MHz, D₂O) δ 0.67 (t, 3H), 1.18 (m,

10H), 1.61 (m, 2H), 4.06 (m, 5H), 4.21 (m, 2H), 4.48 (s, 1H), 4.78

P¹-[(2′,3′)-O-(Phenylethenylmethylidene)uridine-5′-yl]-P⁴-

[uridine-5′-yl]-tetraphosphate (24) and P¹-[(2′,3′)-O-(Phenylethe-

nylmethylidene)uridine-5′-yl]-P⁴-[(2′,3′)-O-(phenylethenylme-

thylidene)uridine-5′-yl]-tetraphosphate (25). Compound 13 and

4Na salt (1.0 g, 1.14 mmol) were dissolved in triﬂuoroacetic acid

(3 mL) and trans-cinnamaldehyde (2.00 mL, 15.9 mmol) added.

The reaction was allowed to proceed at ambient temperature for

1 h. The TFA was removed at <30 °C in vacuo and the residue

partitioned between EtOAc and water, along with the addition of

sufﬁcient concentrated NH₄OH to render pH 7–8. The products in

the aqueous layer were separated via preparative HPLC. The

isolated yield of the monoacetal (24) was 396 mg (38%) and of

the diacetal (25) 112 mg (9.7%). 24: ¹H NMR (300 MHz, D₂O) δ

4.13 (m, 4H), 4.22 (m, 3H), 4.44 (s, 0.5H), 4.94 (d, 2H), 5.59 (d,

J ) 6.6 Hz, 0.4H), 5.73 (d, 6.6 Hz, 0.6H), 5.81 (m, 3H), 6.17 (ddd,

1H), 6.39 (d, 0.3H), 6.85 (dd, 1H), 7.30 (m, 3H), 7.45 (m, 2H),

7.75 (m, 2H). ³¹P NMR (121.47 MHz, D₂O) δ -22.11 (m, 2P),

-10.54 (m, 2P). MS (ES): m/z 903.1 (M - H). 25: ¹H NMR (300

MHz, D₂O) δ 4.14 (m, 4H), 4.39 (m, 1H), 4.86 (m, 4H), 5.56 (d,

J ) 6.6 Hz, 0.8H), 5.65 (t, J ) 6.9 Hz, 1.2H), 5.78 (m, 4H), 6.11

(m, 2H), 5.04 (t, 1H), 5.77 (m, 4H), 7.69 (d, 1H), 7.79 (d, 1H). ³¹

P

NMR (121.47 MHz, D₂O) δ -21.93 (m, 2P), -10.65 (m, 1P),

-10.22 (m 1P). MS (ES): m/z 899.3 (M - H). 31: ¹H NMR (300

MHz, D₂O) δ 0.69 (t, 6H), 1.20 (m, 20H), 1.61 (m, 4H), 4.05 (m,

4H), 4.47 (s, 2H), 4.78 (m, 4H), 5.04 (t, 2H), 5.76 (m, 4H), 7.71

(d, 2H). ³¹P NMR (121.47 MHz, D₂O) δ -21.92 (m, 2P), -10.63

(m, 2P). MS (ES): m/z 1009.6 (M - H).

P¹-[(2′,3′)-O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[cytidine-

5′-yl]-tetraphosphate (33), P¹-[Uridine-5′-yl]-P⁴-[(2′,3′)-O-(2-

phenylethylidene)cytidine-5′-yl]-tetraphosphate (32), and P¹-

[(2′,3′)-O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[(2′,3′)-O-(2-

phenylethylidene)cytidine-5′-yl]-tetraphosphate (34). Compound

18 and 4Na salt (100 mg, 0.114 mmol) were dissolved in 99%

formic acid (1 mL) and phenylacetaldehyde dimethyl acetal (57

µL, 0.342 mmol) added. The reaction was stirred for 16 h, then

the formic acid removed by evaporation. The residue was partitioned

between EtOAc and 1 M NaHCO₃and the products in the aqueous

1020 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

layer puriﬁed via preparative HPLC. The yield of U-side monoacetal

(33) was 26 mg (25%), the C-side monoacetal (32) was 20 mg

(20%), and the diacetal (34) was 21 mg (19%). 33: ¹H NMR (300

MHz, D₂O) δ 2.95 (d, 2H), 4.09 (m, 7H), 4.28 (m, 1H), 4.64 (m,

1H), 4.73 (m, 1H), 5.28 (t, 1H), 5.32 (d, 1H), 5.71 (m, 2H), 6.08

(d, 1H), 7.17 (m, 5H), 7.57 (d, 1H), 7.95 (d, 1H). ³¹P NMR (121.47

MHz, D₂O) δ -22.07 (m, 2P), -10.85 (m, 1P), -10.17 (m, 1P).

1H), 7.20 (m, 5H), 8.08 (s, 1H), 8.20 (s, 1H). ³¹P NMR (121.47

MHz, D₂O) δ -10.31 (d, 1P), -9.52 (d, 1P). MS (ES): m/z 528.1

(M - H).

((2R,3R,4S,5R)-3,4-Dihydroxy-5-(6-(3-phenylureido)-9H-pu-

rin-9-yl)tetrahydrofuran-2-yl)methyl-trihydrogen Diphosphate

(39). Compound 2 and 1NBu₃salt (2, 1.0g, 1.63 mmol) were

dissolved in dry DMF (15 mL) and phenylisocyanate (540 µL, 4.9

mmol) added. The reaction was heated for 1 h at 45 °C. The solvent

was removed and the residue partitioned between EtOAc and 0.25

M NaHCO₃(5 mL). The layers were separated and the product in

the aqueous layer isolated by reverse-phase preparative HPLC. The

1

MS (ES): m/z 890.3 (M - H). 32: H NMR (300 MHz, D₂O) δ

2.95 (d, 2H), 4.07 (m, 7H), 4.42 (m, 1H), 4.69 (m, 2H), 5.23 (d,

2H), 5.75 (m, 2H), 6.04 (d, 1H), 7.15 (m, 5H), 7.73 (dd, 2H). ³¹P

NMR (121.47 MHz, D₂O) δ -22.05 (m, 2P), -11.02 (m, 1P),

-10.28 (m, 1P). MS (ES): m/z 890.3 (M - H). 34: ¹H NMR (300

MHz, D₂O) δ 2.93 (d, 4H), 3.94 (m, 4H), 4.24 (s, 1H), 4.35 (s,

1H), 4.69 (m, 4H), 5.22 (m, 3H), 5.32 (d, 1H), 5.68 (d, 1H), 6.01

(d, 1H), 7.17 (m, 10H), 7.56 (d, 1H), 7.67 (d, 1H). ³¹P NMR (121.47

MHz, D₂O) δ -22.11 (m, 2P), -10.88 (m, 2P). MS (ES): m/z 992.4

(M - H).

1

yield of 39 was 431 mg (48%). 39: H NMR (300 MHz, D₂O) δ

4.09 (m, 2H), 4.21 (m, 1H), 4.36 (t, 1H), 4.48 (t, 1H), 5.78 (d,

1H), 6.54 (t, 1H), 6.77 (m, 4H), 8.18 (s, 1H), 8.25 (s, 1H). ³¹P

NMR (121.47 MHz, D₂O) δ -10.08 (d, 1P), -8.76 (d, 1P). MS

(ES): m/z 545.1 (M - H).

((3aS,4R,6R,6aS)-2-Benzyl-6-(6-(3-phenylureido)-9H-purin-9-

yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-trihydrogen Di-

phosphate (41). Compound 40, 1NH₄salt (200 mg, 0.355 mmol),

and tributylamine (200 µL, 0.841 mmol) were dissolved in dry DMF

(3 mL). Phenylisocyanate (194 µL, 1.78 mmol) was added and the

reaction mixture heated for 60 h at 30 °C. The solvent was removed

and the residue partitioned between EtOAc and 0.2 M NaHCO₃(3

mL). The product in the aqueous layer was isolated by reverse-

phase preparative HPLC. The yield of the title compound (41) was

31 mg (13.5%). Compound 41 proved to be somewhat unstable,

slowly degrading to the corresponding monophosphate (56).

Consequently, heating was avoided during the isolation of this

molecule, and it was stored at -80 °C immediately after the

lyophilization step. When handled appropriately, the typical content

of compound 56 was 3–5%. 41: ¹H NMR (300 MHz, D₂O) δ 3.06

(d, 2H), 3.94 (m, 2H), 4.44 (s, 1H), 4.97 (d,1H), 5.13 (m, 1H),

5.35 (t, 1H), 5.71 (d, J ) 3.3 Hz, 1H), 6.89 (t, 1H), 7.10 (t, 2H),

P¹-[(2′,3′)-O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[uridine-5′-

yl]-triphosphate (35) and P¹-[(2′,3′)-O-(2-Phenylethylidene)-

uridine-5′-yl]-P⁴-[(2′,3′)-O-(2-phenylethylidene)uridine-5′-yl]-

triphosphate (36). Compound 19 and 3Na salt (100 mg, 0.129

mmol) were dissolved in 99% formic acid (1 mL) and phenylac-

etaldehyde dimethyl acetal (64 µL, 0.386 mmol) added. The reaction

was stirred for 16 h, and the formic acid then removed by

evaporation. The residue was partitioned between EtOAc and 0.5

M NaHCO₃and the products in the aqueous layer puriﬁed via

preparative HPLC. The yield of the monoacetal (35) was 40 mg

1

(38%) and of the diacetal (36) 24 mg (20%). 35: H NMR (300

MHz, D₂O) δ 2.93 (d, 2H), 3.98 (d, 5H), 4.16 (d, 3H), 4.67 (m,

2H), 5.23 (t, 1H), 5.31 (s, 1H), 5.67 (d, 1H), 5.74 (m, 2H), 7.16

(m, 5H), 7.55 (d, 1H), 7.73 (d, 1H). ³¹P NMR (121.47 MHz, D₂O)

δ -21.99 (m, 1P), -10.80 (m, 1P) 10.30 (m, 1P). MS (ES): m/z

1

811.6 (M - H). 36: H NMR (300 MHz, D₂O) δ 2.91 (d, 4H),

1

3.89 (s, 4H), 4.13 (s, 2H), 4.63 (m, 2H), 5.21 (t, 2H), 5.32 (d, 2H),

5.65 (d, 2H), 7.15 (m, 10H), 7.49 (d, 2H). ³¹P NMR (121.47 MHz,

D₂O) δ -21.99 (m, 1P), -10.62 (m, 2P). MS (ES): m/z 913.3 (M

- H).

7.21 (m, 7H), 8.31 (s, 1H), 8.44 (s, 1H).). H NMR (300 MHz,

MeOD + D₂O) δ 3.14 (d, 2H), 4.16 (m, 2H), 4.54 (s, 1H), 5.15 (d,

1H), 5.33 (m, 2H), 6.17 (d, J ) 3.3 Hz, 1H), 7.12 (t, 1H), 7.31 (m,

7H), 7.61 (d, 2H), 8.70 (s, 1H), 8.73 (s, 1H). ³¹P NMR (121.47

MHz, D₂O) δ -10.44 (d, 1P), -9.52 (d, 1P). MS (ES): m/z 648.8

(M - H).

P¹-[(2′,3′)-O-(2-Phenylethylidene)adenosine-5′-yl]-P⁴-[adenos-

ine-5′-yl]-tetraphosphate (37) and P¹-[(2′,3′)-O-(2-Phenyleth-

ylidene)adenosine-5′-yl]-P⁴-[(2′,3′)-O-(2-phenylethylidene)ade-

nosine-5′-yl]-tetraphosphate (38). The free acid of 11 (500 mg,

0.598 mmol) was dissolved in 99% formic acid (5 mL) and

phenylacetaldehyde dimethyl acetal (297 µL, 1.79 mmol) added.

The reaction was stirred for 21 h, then the formic acid removed by

evaporation. The residue was partitioned between EtOAc and 1 M

NaHCO₃and the products in the aqueous layer puriﬁed via

preparative HPLC. The yield of monoacetal (37, 48506) was 155

mg (27.6%) and of the diacetal (38, 48808) 38 mg (6.1%). 37: ¹H

NMR (300 MHz, D₂O) δ 3.02 (d, 2H), 4.05 (m, 2H), 4.14 (m,

2H), 4.22 (m, 1H), 4.40 (m, 2H), 4.55 (t, 1H), 4.92 (m, 1H), 5.04

(m, 1H), 5.34 (t, 1H), 5.68 (d, 1H), 5.87 (d, 1H), 7.24 (m, 5H),

8.03 (d, 1H), 8.19 (s, 1H), 8.31 (s, 1H). ³¹P NMR (121.47 MHz,

D₂O) δ -21.74 (m, 2P), -10.35 (m, 1P), -9.98 (m, 1P). MS (ES):

m/z 937.5 (M - H). 38: ¹H NMR (300 MHz, D₂O) δ 2.95 (d, 4H),

4.01 (m, 4H), 4.30 (m, 2H), 4.87 (m, 2H), 4.96 (m, 2H), 5.23 (t,

2H), 5.61 (d, 2H), 7.15 (m, 10H), 7.92 (s, 2H), 8.18 (s, 2H). ³¹P

NMR (121.47 MHz, D₂O) δ -21.84 (m, 2P), -10.40 (m, 2P). MS

(ES): m/z 1039.8 (M - H).

P¹-[(2′,3′)-O-(2-Phenylethylidene)-6-(3-phenylureido)adenos-

ine-5′-yl]-P⁴-[(2′,3′)-O-(2-phenylethylidene)-6-(3-phenylureido)ad-

enosine-5′-yl]-tetraphosphate (42). Compound 41 (60 mg, 0.092

mmol) was dissolved in dry DMF (1.0 mL) and NBu₃(100 µL,

0.42 mmol). 1,1′-Carbonyldiimidazole (20 mg, 0.123 mmol) was

added and the reaction mixture allowed to stir overnight at ambient

temperature. The reaction was quenched with water (1 mL) and 1

M NaHCO₃(1 mL). The desired product was isolated by reverse-

phase preparative HPLC. The yield of the title compound (42) was

1

9 mg (16%). 42: H NMR (300 MHz, D₂O + acetonitrile d3) δ

3.04 (t, 4H), 4.01 (s, 4H), 4.41 (s, 2H), 4.87 (d, 2H), 4.99 (t, 2H),

5.22 (t, 2H), 5.95 (d, 2H), 7.08 (t, 2H), 7.26 (m, 14H), 7.51 (d,

4H), 8.39 (s, 2H), 8.44 (s, 2H). ³¹P NMR (121.47 MHz, D₂O) δ

-21.31 (m, 2P), -10.13 (m, 2P). MS (ES): m/z 1279.9 (M - H).

((2R,3R,4S,5R)-5-(4-(3-(4-Fluorophenyl)ureido)-2-oxopyrimi-

din-1(2H)-yl)-3,4-dihydroxytetrahydrofuran-2-yl)methyl-dihy-

drogen Phosphate (44). Compound 43 and 1 NBu₃salt (51 mg,

0.1 mmol) were dissolved in dry DMF (0.5 mL) and p-ﬂuorophe-

nylisocyanate (34 µL, 0.3 mmol) added. The reaction was heated

at 60 °C for 5 min, then at 30 °C for 1 h. The solvent was removed

by evaporation under vacuum and the residue partitioned between

EtOAc (10 mL) and 0.5 M NaHCO₃. The product in the aqueous

layer was isolated by reverse-phase preparative HPLC. The yield

((3aS,4R,6R,6aS)-6-(6-Amino-9H-purin-9-yl)-2-benzyltetrahy-

drofuro[3,4-d][1,3]dioxol-4-yl)methyl-trihydrogen Diphosphate

(40). Compound 2 and 1Na salt (800 mg, 1.78 mmol) were

dissolved in 99% formic acid (5 mL), and phenylacetaldehyde

dimethyl acetal (589 µL, 3.56 mmol) was added. The mixture was

stirred overnight at 35 °C and evaporated to dryness. The residue

was partitioned between EtOAc and 1 M NaHCO₃. The resultant

partial emulsion was centrifuged (4000 rpm) and the layers

separated. The aqueous layer containing the product was concen-

trated to 20 mL and the product puriﬁed by reverse-phase

preparative HPLC. The yield of the title compound (40) was 252

mg (27%). 40: ¹H NMR (300 MHz, D₂O) δ 3.03 (d, 2H), 3.94 (d,

2H), 4.40 (s, 1H), 4.92 (d, 1H), 5.14 (m, 1H), 5.33 (t, 1H), 5.73 (d,

1

of the title compound (44) was 51 mg (80%). 44: H NMR (300

MHz, D₂O) δ 4.05 (m, 5H), 5.78 (d, 1H), 6.24 (d, 1H), 6.97 (t,

2H), 7.28 (m, 2H), 8.14 (d, 1H). ³¹P NMR (121.47 MHz, D₂O) δ

1.39 (s). MS (ES): m/z 461.4 (M - H).

((3aS,4R,6R,6aS)-6-(4-Amino-2-oxopyrimidin-1(2H)-yl)-2-

benzyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydro-

gen Phosphate (45). Compound 43 (1.0g, 3.09 mmol), in the free

acid form, was dissolved in 99% formic acid (7.5 mL) and

phenylacetaldehyde dimethyl acetal (1.03 mL, 6.2 mmol) added.

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1021

The reaction was stirred overnight at ambient temperature, after

which the volatiles were removed under vacuum. The residue was

partitioned between EtOAc and 1 M NaHCO₃, and the product in

the aqueous layer was isolated by reverse-phase preparative HPLC.

4H), 5.17 (t, 1H), 5.28 (t, 1H), 5.35 (d, 1H), 5.41 (d, 1H), 5.68 (d,

1H), 6.24 (d, 1H), 6.98 (t, 2H), 7.18 (m, 10H), 7.31 (m, 2H), 7.53

(d, 1H), 7.90 (d, 1H).). ³¹P NMR (121.47 MHz, D₂O) δ -21.57

(m, 2P), -10.62 (m, 2P). MS (ES): m/z 1129.7 (M - H).

1

The yield of the title compound (45) was 650 mg (49%). 45: H

((3aS,4R,6R,6aS)-6-(6-Amino-9H-purin-9-yl)-2-benzyltetrahy-

drofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phosphate

(52). The free acid of 1 (10.0g, 28.8 mmol) was dissolved in

triﬂuoroacetic acid (50 mL) and cooled to 0–5 °C followed by the

addition of phenylacetaldehyde dimethyl acetal (13.5 mL, 88.5

mmol). Three additional portions of phenylacetaldehyde dimethyl

acetal (2.5 mL, 15mmol) were added at 2 h, 2.75 h, and 3.5 h. The

reaction was stirred for a total of 7 h, and the volatiles were removed

in vacuo at <40 °C. The residue was partitioned between EtOAc

and 0.75 M NaHCO₃. The aqueous layer was back extracted with

EtOAc, and the product in the aqueous layer was isolated by

reverse-phase preparative HPLC. The yield of the title compound

NMR (300 MHz, D₂O) δ 2.98 (d, 2H), 3.77 (m, 2H, 4.17 (s, 1H),

4.71 (m, 2H), 5.28 (t, 1H), 5.52 (d, 1H), 5.86 (d, 1H), 7.20 (m,

5H), 7.63 (d, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 2.87 (s). MS

(ES): m/z 424.2 (M - H).

((3aS,4R,6R,6aS)-2-Benzyl-6-(4-(3-(4-ﬂuorophenyl)ureido)-2-

oxopyrimidin-1(2H)-yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)-

methyl-hydrogen Phosphate (46). Compound 45 (100 mg, 0.218

mmol) and NBu₃(104 µL, 0.436 mmol) were dissolved in dry DMF

(2 mL). 4-Fluorophenylisocyanate (99 µL, 0.871 mmol) was added

and the reaction stirred at room temperature. At 1 h, an additional

portion of 4-ﬂuorophenylisocyanate (99 µL, 0.871 mmol) was

added. After 3 h, the solvent was removed and the residue

partitioned between EtOAc and 1 M NaHCO₃. The product in the

aqueous layer was isolated by reverse-phase preparative HPLC. The

yield of the title compound (46) was 102 mg (83%). 46: ¹H NMR

(300 MHz, D₂O) δ 3.00 (d, 2H), 3.81 (m, 2H), 4.34 (m, 1H), 4.72

(m, 2H), 5.28 (t, 1H), 5.44 (d, 1H), 6.20 (d, 1H), 6.99 (t, 2H), 7.27

(m, 2H), 7.92 (d, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 2.39 (s).

MS (ES): m/z 563.2 (M - H).

1

(52) was 7.5g (59%). 52: H NMR (300 MHz, D₂O) δ 2.99 (d,

2H), 3.80(m, 2H), 4.32 (s, 1H), 4.82 (d, 1H), 5.07 (m, 1H), 5.28 (t,

1H), 5.72 (d, 1H), 7.17 (m, 5H), 7.99 (s, 1H), 8.09 (s, 1H). ³¹P

NMR (121.47 MHz, D₂O) δ 1.45 (s). MS (ES): m/z 448.3 (M -

H). HPLC retention time (C₁₈, ion pairing): 10.64 min (100%).

((3aS,4R,6R,6aS)-6-(6-Amino-9H-purin-9-yl)-2-phenyltetrahy-

drofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phosphate

(53). Compound 53 was prepared from 1 (free acid) according to

the method for the synthesis of 52, substituting benzaldehyde (3

equiv) for phenylacetaldehyde dimethyacetal. The yield of the title

compound (53) was 92%, following reverse-phase preparative

HPLC puriﬁcation. The product consisted of a 50:50 ratio of acetal

diastereomers, as determined by HPLC and NMR (MeOD). 53:

¹H NMR (300 MHz, MeOD) δ 4.07 (m, 2H), 4.56 (m, 0.5H), 4.66

(m, 0.5H), 5.24 (m, 1H), 5.46 (m, 1H),6.00 (s, 0.5H), 6.24 (s, 0.5H),

6.36 (d, J ) 3.0 Hz, 0.5H), 6.39 (d, J ) 3.3 Hz, 0.5H), 7.39 (m,

3H), 7.52 (m, 1H), 7.58 (m, 1H), 8.20 (d, 1H), 8.49 (s, 0.5H), 8.56

(s, 0.5H). ³¹P NMR (121.47 MHz, D₂O) δ 1.58 (s 0.5P), 1.70 (s,

0.5P). MS (ES): m/z 434.5 (M - H). HPLC retention times (C₁₈,

ion pairing): 9.99 min (50%), 10.17 (50%).

P¹-[Uridine-5′-yl]-P⁴-[4-(3-(4-ﬂuorophenyl)ureido)cytidine-5′-

yl]-tetraphosphate (47). Compound 6 and 2NBu₃salt (8.93 g, 10.5

mmol) were dissolved in dry DMF (total solution volume 50 mL)

and DCC (2.04 g, 9.9 mmol) added. The reaction was stirred for

1.25 h at 30 °C, during which time it became heterogeneous because

of precipitated dicyclohexylurea. ³¹P NMR showed complete

conversion to the cyclic trimeta phosphate. A solution of CMP 6-(p-

ﬂuoro)phenylurea and 1NBu₃salt (44, 3.38 g, 5.25 mmol) in dry

DMF (50 mL) was added, and the solution was kept at 30 °C for

3 days. Water (10 mL) was added, and the reaction mixture was

stirred for 30 min, after which the solvent was removed by

evaporation. The residue was dissolved in water (100 mL), and

the precipitated dicyclohexylurea was removed by ﬁltration. The

ﬁltrate was concentrated and the product isolated by reverse-phase

preparative HPLC. The yield of the title compound (47) was 2.12 g

(41%, based on 44). 47: ¹H NMR (300 MHz, D₂O) δ 4.11 (m,

10H), 5.74 (m, 3H), 6.30 (d, 1H), 6.97 (t, 2H), 7.29 (m, 2H), 7.71

(d, 1H), 8.15 (d, 1H). ³¹P NMR (121.47 MHz, D₂O) δ -21.67 (m,

2P), -10.32 (m, 2P). MS (ES): m/z 925.4 (M - H).

((3aS,4R,6R,6aS)-6-(6-Amino-9H-purin-9-yl)-2-styryltetrahy-

drofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phosphate

(54). Compound 54 was prepared from 1 (free acid) according to

the method for the synthesis of 52, substituting trans-cinnamalde-

hyde (2 equiv) for phenylacetaldehyde dimethyl acetal. Following

overnight stirring at 0–5 °C, further portions of trans-cinnamalde-

hyde (4 × 0.9 equiv) were added over a 2 h period, to enhance the

conversion to product. The yield of the title compound (54) was

51%, following preparative HPLC puriﬁcation. The product con-

sisted of a 66:33 ratio of acetal diastereomers, as determined by

HPLC and NMR (MeOD). 54: ¹H NMR (300 MHz, MeOD) δ 4.08

(m, 2H), 4.50 (m, 0.66H), 4.62 (m, 0.33H), 5.11 (dd, 0.33H), 5.18

(dd, 0.66H), 5.37 (m, 1H), 5.65 (d, J ) 6.9 Hz, 0.33H), 5.89 (d, J

) 6.3 Hz, 0.66H), 6.29 (m, 2H), 6.87 (d, J ) 15.9 Hz, 1H), 7.33

(d, 3H), 7.47 (m, 2H), 8.22 (s, 1H), 8.49 (s, 0.66H), 8.56 (s, 0.33H).

³¹P NMR (121.47 MHz, D₂O) δ 1.17 (s 0.33P), 1.28 (s, 0.66P).

MS (ES): m/z 460.6 (M - H) HPLC retention times (C₁₈, ion

pairing): 10.96 min (66%), 11.11 (33%).

P¹-[(2′,3′)-O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[4-(3-(4-

ﬂuorophenyl)ureido)cytidine-5′-yl]-tetraphosphate (48), P¹-[Uri-

dine-5′-yl]-P⁴-[(2′,3′)-O-(2-phenylethylidene)-4-(3-(4-ﬂuorophe-

nyl)ureido)cytidine-5′-yl]-tetraphosphate (49), and P¹-[(2′,3′)-

O-(2-Phenylethylidene)uridine-5′-yl]-P⁴-[(2′,3′)-O-(2-

phenylethylidene)-4-(3-(4-ﬂuorophenyl)ureido)cytidine-5′-yl]-

tetraphosphate (50). Compound 47 (100 mg, 0.108 mmol), in the

free acid form, was dissolved in 99% formic acid (2 mL) and

phenylacetaldehyde dimethyl acetal (107 µL, 0.65 mmol) added.

The reaction was heated at 30 °C for 45 min, after which the formic

acid was removed at ambient temperature. The residue was

partitioned between EtOAc and 0.8 M NaHCO₃. The products in

the aqueous layer were puriﬁed by reverse-phase preparative HPLC.

In addition to the recovered starting material (47, 38 mg), the yields

were 39 mg of the uridine-side monoacetal (48, 32%), 14 mg of

the cytidine-side monoacetal (49, 12%), and 15 mg of the diacetal

(50, 12%). 48: ¹H NMR (300 MHz, D₂O) δ 2.86 (d, 2H), 4.10 (m,

8H), 4.66 (m, 2H), 5.19 (t, 1H), 5.31 (d, 1H), 5.69 (d, 1H), 5.78

(d, 1H), 6.30 (d, 1H), 6.97 (t, 2H), 7.13 (m, 5H), 7.30 (m, 2H),

7.54 (d, 1H), 8.16 (d, 1H). ³¹P NMR (121.47 MHz, D₂O) δ -21.71

(m, 2P), -10.62 (m, 1P), -9.92 (m, 1P). MS (ES): m/z 1027.2 (M

- H). 49: ¹H NMR (300 MHz, D₂O) δ 2.98 (d, 2H), 4.19 (m, 7H),

4.42 (m, 1H), 4.68 (m, 2H), 5.26 (t, 1H), 5.37 (d, 1H), 5.66 (d,

1H), 5.75 (d, 1H), 6.24 (d, 1H), 6.96 (t, 2H), 7.21 (m, 5H), 7.30

(d, 2H), 7.61 (d, 1H), 7.88 (d, 1H). ³¹P NMR (121.47 MHz, D₂O)

δ -21.62 (m, 2P), -10.60 (m, 1P), -10.19 (m, 1P). MS (ES): m/z

((3aS,4R,6R,6aS)-6-(6-Amino-9H-purin-9-yl)-2-(phenylethy-

nyl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phos-

phate (55). Compound 55 was prepared from 1 (free acid) according

to the method for the synthesis of 52, substituting phenylpropargyl

aldehyde (3 equiv) for phenylacetaldehyde dimethyl acetal. The

reaction was stirred for three days at RT, giving a 36% yield of

the title compound (55) following preparative HPLC puriﬁcation.

The product consisted of a 35:65 ratio of acetal diastereomers, as

determined by HPLC and NMR (MeOD). 55: ¹H NMR (300 MHz,

MeOD) δ 4.09 (m, 2H), 4.56 (m, 0.33H), 4.76 (m, 0.66H), 5.19

(dd, 0.66H), 5.30 (dd, 0.33H), 5.42 (dd, 0.65H), 5.51 (dd, 0.35H),

6.08 (s, 0.65H), 6.29 (d, J ) 3.3 Hz, 0.33H), 6.31 (s, 0.35H), 6.55

(d, J ) 3.0 Hz, 0.66H), 7.38 (m, 4H), 7.54 (m, 1H), 8.22 (d, 1H),

8.51 (d, 1H) ³¹P NMR (121.47 MHz, D₂O) δ 1.35 (s). (MS (ES):

m/z 458.4 (M - H) HPLC retention times (C₁₈, ion pairing): 10.89

min (35%), 11.09 (65%).

1

1027.2 (M - H). 50: H NMR (300 MHz, D₂O) δ 2.88 (d, 2H),

3.00 (d, 2H), 4.00 (m, 4H), 4.18 (m, 1H), 4.36 (m, 1H), 4.68 (m,

1022 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

((2R,3R,4S,5R)-3,4-Dihydroxy-5-(6-(3-phenylureido)-9H-pu-

rin-9-yl)tetrahydrofuran-2-yl)methyl-dihydrogen Phosphate

(51). Compound 51 was synthesized from 1 (NBu₃salt) according

to the method for the synthesis of 39, using phenyl isocyanate (5

equiv). The yield of the title compound was 17%. The low yield

of 51 was attributable to byproducts arising from the competitive

acylation of the hydroxyl groups. 51: ¹H NMR (300 MHz, D₂O) δ

3.99 (, 3H), 4.23 (s, 1H), 4.34 (t, 1H), 4.58 (t, 1H), 5.97 (d, J )

5.4 Hz, 1H), 6.90 (t, 1H), 7.10 (t, 2H), 7.18 (d, 2H), 8.40 (s, 1H),

8.42 (s, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 1.65 (s).

cyanate (12 equiv) for n-hexylisocyanate. The isocyanate was added

in 4 equal portions, with heating at 50 °C over 6 days. The reaction

was concentrated to an oil and treated with 1 M NaHCO₃to cleave

the transient acylphosphate species. This solution was partitioned

between aqueous NaHCO₃and ether, and the products in the

aqueous layer were separated by reverse-phase preparative HPLC.

The yield of 59 was 10%. 59: ¹H NMR (300 MHz, DMSO-d₆)

1.54 (m, 4H), 1.67 (m, 2H), 1.88 (m, 2H), 3.04 (d, 2H), 3.75 (m,

2H), 4.06 (m, 1H), 4.38 (s, 1H), 4.96 (d, 1H), 5.24 (m, 2H), 6.15

(d, 1H), 7.23 (m, 5H), 8.53 (s, 1H), 8.74 (s, 1H), 9.4 (d, 1H), 9.60

(broad s, 1H). δ ³¹P NMR (121.47 MHz, D₂O) δ 1.49 (s). MS

(ES): m/z 559.3 (M - H). HPLC retention time (C₁₈, ion pairing):

13.50 min (100%).

((3aS,4R,6R,6aS)-2-Benzyl-6-(6-(3-phenylureido)-9H-purin-9-

yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phos-

phate (56). Compound 52 and 1 NH₄salt (5.0g, 10.7 mmol) were

dissolved in DMF (20 mL) and tributylamine (5.11 mL, 21.4 mmol)

added, stirred for 5 min, and then evaporated to an oil. This oil

was dissolved in DMF (35 mL), and phenyl isocyanate (3.51 mL.

32.2 mmol) was added and the reaction heated to 30–35 °C

overnight. The volatiles were evaporated, and the residue was

partitioned between BuOAc (100 mL), toluene (200 mL), and water

(300 mL). The aqueous layer was separated, concentrated, and the

product puriﬁed by reverse-phase preparative HPLC. The puriﬁed

yield of the title compound 56 following lyophilization was 3.3g

((3aS,4R,6R,6aS)-6-(6-(3-Ethylureido)-9H-purin-9-yl)-2-phe-

nyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phos-

phate (60). Compound 60 was prepared from 53 according to the

method for the synthesis of 57, substituting ethylisocyanate (6 equiv)

for n-hexylisocyanate. The reaction was heated at 35–40 °C for 4

days, with daily additions of the isocyanate (1.5 equiv × 4) over

this time. The conversion by this time was around 92%. Following

alkaline treatment to cleave the transient acylphosphate species,

workup (aqueous NaHCO₃/ether), and preparative HPLC, the yield

of 60 was 86%. The product consisted of an approximately 60:40

ratio of acetal diastereomers (60:40 by NMR (MeOD); 56:44 by

1

(54%). 56: H NMR (300 MHz, MeOD) δ 3.14 (d, J ) 4.5 Hz,

2H), 4.01 (m, 2H), 4.53 (s, 1H), 5.07 (d, 1H), 5.31 (m, 2H), 6.19

(d, J ) 3.3 Hz, 1H), 7.11 (t, 1H), 7.29 (m, 7H), 7.61 (d, 2H), 8.69

(s, 1H), 8.81 (s, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 3.15 (s).

MS (ES): m/z 567.2 (M - H). HPLC retention time (C₁₈, ion

pairing): 13.48 min (100%).

1

HPLC). 60: H NMR (300 MHz, MeOD) δ 1.26 (t, 3H), 3.41 (q,

2H), 4.06 (m, 2H), 4.61 (m, 0.6H), 4.70 (s, 0.4H), 5.28 (m, 1H),

5.51 (m, 1H), 6.02 (s, 0.4H), 6.27 (s, 0.6H), 6.47 (t, 1H), 7.40 (m,

3H), 7.52 (m, 1H), 7.59 (m, 1H), 8.56 (d, 1H), 8.72 (s, 0.6H), 8.81

(s, 0.4H). ³¹P NMR (121.47 MHz, D₂O) δ 1.39 (s, 0.5P), 1.54 (s,

0.5P). MS (ES): m/z 505.2 (M - H). HPLC retention time (C₁₈,

ion pairing): 11.58 min (56%), 11.95 min 44%).

((3aS,4R,6R,6aS)-2-Benzyl-6-(6-(3-hexylureido)-9H-purin-9-

yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phos-

phate (57). Compound 52 and 1 NH₄salt (8.0g, 17.2 mmol) were

dissolved in DMF (50 mL) and tributylamine (8.17 mL, 34.3 mmol)

and the solution was concentrated to approximately one-half volume

under vacuum to remove ammonia. Further DMF (25 mL) was

added, after which n-hexylisocyanate (7.0 mL, 48.3 mmol) was

added over 2 h, with vigorous stirring and at such a rate as to avoid

solidiﬁcation of the reaction mixture. The reaction was heated for

3 days at 35 °C and was concentrated to an oil. This residue was

reconstituted in 1 M NaHCO₃(20 mL) and water (300 mL), and

the resulting suspension heated for 10 min at 50 °C in order to

cleave any acylphosphate intermediates. The mixture was cooled,

water (100 mL) and ether (200 mL) were added, and the layers

were separated. The product containing the aqueous layer was

concentrated to around 100 mL and an equal volume of acetonitrile

added as an aid to solubility. The product was isolated by reverse-

phase preparative HPLC. The yield of 57 was 5.2 g (52%). 57: ¹H

NMR (300 MHz, D₂O) δ 0.76 (t, 3H), 1.17 (s, 6H), 1.38 (m, 2H),

2.94 (d, 2H), 3.09 (t, 2H), 3.80 (m, 2H), 4.38 (s, 1H), 4.85 (d, 1H),

5.07 (t, 1H), 5.25 (t, 1H), 5.82 (d, 1H), 7.12 (m, 5H), 8.26 (s, 1H),

8.44 (s, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 5.12 (s). MS (ES):

m/z 575.3 (M - H). HPLC retention time (C₁₈, ion pairing): 14.90

min (100%).

((3aS,4R,6R,6aS)-6-(6-(3-Ethylureido)-9H-purin-9-yl)-2-

styryltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen

Phosphate (61). Compound 61 was prepared from 54 according

to the method for the synthesis of 57, substituting ethylisocyanate

(7 equiv) for n-hexylisocyanate. The isocyanate was added in daily

portions (1 equiv × 7), with heating at 30–35 °C over 7 days. The

conversion by this time was around 65%. Following alkaline

treatment to cleave the transient acylphosphate species, workup

(aqueous NaHCO₃/ether), and preparative HPLC, the yield of 61

was 50%. The product consisted of an approximately 66:33 ratio

of acetal diastereomers (65:35 by NMR (MeOD); 66:33 by HPLC).

61: ¹H NMR (300 MHz, MeOD) δ 1.26 (t, 3H), 3.41 (q, 2H), 4.07

(m, 2H), 4.54 (m, 0.65H), 4.65 (s, 0.35H), 5.14 (dd, 0.35H), 5.21

(dd, 0.65H), 5.38 (m, 1H), 5.65 (d, J ) 6.6 Hz, 0.35H), 5.90 (d, J

) 6.3 Hz, 0.65H), 6.21 (dd, J ) 6.3 Hz, 0.65H), 6.39 (m, 1.35H),

6.89 (dd, 1H), 7.31 (m, 3H), 7.48 (m, 2H), 8.57 (d, 1H), 8.69 (s,

0.6H), 8.77 (s, 0.4H). ³¹P NMR (121.47 MHz, D₂O) δ 1.90 (s,

0.33P), 1.98 (s, 0.66P). MS (ES): m/z 531.2 (M - H). HPLC

retention times (C₁₈, ion pairing): 12.06 min (66%), 12.39 (33%).

((2S,3aS,4R,6R,6aS)-6-(6-(3-Ethylureido)-9H-purin-9-yl)-2-

styryltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen

Phosphate (62). Compound 61 (50 mg; a 66% trans:33% cis

mixture of diastereomers) was dissolved in 25% aqueous acetic

acid (2 mL) and was warmed at 35 °C for 1.5 h. Ion pairing HPLC

(see General Methods) indicated that the less stable cis-acetal isomer

was completely cleaved by this time, leaving a mixture of the trans

diastereomer, the 2′,3′-diol arising from the cleavage of the cis

diastereomer, and cinnamaldehyde. The products were separated

and puriﬁed via preparative HPLC. Following lyophilization, the

yield of title compound 62 was 23 mg (69.7%). 62: ¹H NMR (300

MHz, MeOD) δ 1.26 (t, 3H), 3.41 (q, 2H), 4.03 (m, 2H), 4.56 (s,

1H), 5.24 (dd, 1H), 5.45 (dd, 1H), 5.91 (d, J ) 6.0 Hz, 1H), 6.22

(dd, J ) 6.0 Hz, J ) 15.9 Hz, 1H), 6.40 (d, J ) 3.3 Hz, 1H), 6.87

(d, J ) 16.2 Hz, 1H), 7.29 (m, 3H), 7.46 (d, 2H), 8.56 (s, 1H),

8.92 (s, 1H). ³¹P NMR (121.47 MHz, D₂O) δ 1.29 (s). MS (ES):

m/z 531.2 (M - H). HPLC retention time (C₁₈, ion pairing): 12.12

min (100%).

((3aS,4R,6R,6aS)-2-Benzyl-6-(6-(3-ethylureido)-9H-purin-9-

yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen Phos-

phate (58). Compound 58 was prepared from 52 according to the

method for the synthesis of 57, substituting ethylisocyanate (6.5

equiv) for n-hexylisocyanate. The reaction was heated for a total

of 3 days at 50 °C, then concentrated to an oil. Following alkaline

treatment to cleave the transient acylphosphate species, the residue

was partitioned between aqueous NaHCO₃and ether. The products

in the aqueous layer were separated by reverse-phase preparative

1

HPLC, giving 58 in 47% yield. 58: H NMR (300 MHz, D₂O) δ

1.06 (t, 3H), 2.98 (d, 2H), 3.20 (q, 2H), 3.83 (d, 2H), 4.36 (s, 1H),

4.84 (d, 1H), 5.10 (s, 1H), 5.28 (t, 1H), 5.74 (s, 1H), 7.17 (s, 5H),

8.19 (s, 1H), 8.28 (s, 1H).³¹P NMR (121.47 MHz, D₂O) δ 1.21

(s). MS (ES): m/z 519.2 (M - H). HPLC retention time (C₁₈, ion

pairing): 12.00 min (100%).

((3aS,4R,6R,6aS)-2-Benzyl-6-(6-(3-cyclopentylureido)-9H-pu-

rin-9-yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydro-

gen Phosphate (59). Compound 59 was prepared from 52 according

of the method for the synthesis of 57, substituting cyclopentyliso-

((3aS,4R,6R,6aS)-6-(6-(3-Hexylureido)-9H-purin-9-yl)-2-

styryltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1023

Phosphate (63). Compound 63 was prepared from 54 according

to the method for the synthesis of 57. n-Hexylisocyanate (4.3 equiv)

was added and the reaction heated at 35 °C overnight. A further

portion of n-hexylisocyanate (2.2 equiv) was added, and the reaction

heated an additional 3 days. Following alkaline treatment, workup

(aqueous NaHCO₃/ether), and preparative HPLC, the yield of 63

was 38%. The product consisted of an approximately 75:25 ratio

of acetal diastereomers (80:20 by NMR (D₂O); 75:25 by HPLC).

acknowledge Sanjoy Mahanty and Anna Morgan for performing

some of the platelet aggregation studies.

Supporting Information Available: ¹H NMR (D₂O, 300 MHz)

for compounds 11, 13, 18, 21, 53-56, 59–62, 65, and 66. NOE

(D₂O, 300 MHz) correlations for compounds 21, 31, 38, 45, 58,

and 62. This material is available free of charge via the Internet at

http://pubs.acs.org.

1

63: H NMR (300 MHz, D₂O) δ 0.69 (t, 3H), 1.19 (m, 6H), 1.44

(t, 2H), 3.18 (t, 2H), 3.85 (s, 2H), 4.46 (d, 1H), 5.05 (m, 1H), 5.25

(m, 0.2H), 5.30 (m, 0.8H), 5.60 (d, 0.2H), 5.80 (d, 0.8H), 6.07

(dd, 0.8H)), 6.22 (m, 1.2H), 6.76 (t, 1H), 7.20 (m, 3H), 7.29 (m,

2H), 8.37 (s, 1H), 8.48 (s, 0.8H), 8.57 (s, 0.2H). ³¹P NMR (121.47

MHz, D₂O) δ 4.99(s, 0.2P), 5.05 (s, 0.8P). MS (ES): m/z 587.4 (M

- H). HPLC retention time (C₁₈, ion pairing): 15.43 min (75%),

16.37 (25%).

References

(1) Burnstock, G.; Williams, M. P2 purinergic receptors: modulation of

cell function and therapeutic potential. J. Pharmacol. Exp. Ther. 2000,

295, 862–869.

(2) Jacobson, K. A.; Jarvis, M. F.; Williams, M. Purine and pyrimidine

(P2) receptors as drug targets. J. Med. Chem. 2002, 45, 4057–4093.

(3) Burnstock, G. Potential therapeutic targets in the rapidly expanding

ﬁeld of purinergic signaling. Clin. Med. 2002, 2, 45–53.

((3aS,4R,6R,6aS)-6-(6-(3-Phenylureido)-9H-purin-9-yl)-2-

styryltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydrogen

Phosphate (64). Compound 64 was prepared from 54 according

to the method for the synthesis of 56. The reaction was allowed to

stir at 40 °C for 2 days. Following workup and preparative HPLC,

the yield of the title compound was 22%. The product consisted of

an approximately 65:35 ratio of acetal diastereomers (66:33 by

(4) Abbracchio, M. P.; Burnstock, G.; Boeynaems, J. M.; Barnard, E. A.;

Boyer, J. L.; Kennedy, C.; Knight, G. E.; Fumagalli, M.; Gachet, C.;

Jacobson, K. A.; Weisman, G. A. International Union of Pharmacology

LVIII: update on the P2Y G protein-coupled nucleotide receptors: from

molecular mechanisms and pathophysiology to therapy. Pharmacol.

ReV. 2006, 58, 281–341.

(5) Nichols, K. K.; Yerxa, B.; Kellerman, D. J. Diquafosol tetrasodium:

a novel dry eye therapy. Expert Opin. InVest. Drugs 2004, 13, 47–54.

(6) Yerxa, B. R.; Sabater, J. R.; Davis, C. W.; Stutts, M. J.; Lang-Furr,

M.; Picher, M.; Jones, A. C.; Cowlen, M.; Dougherty, R.; Boyer, J.;

Abraham, W. M.; Boucher, R. C. Pharmacology of INS37217 [P(1)-

(uridine 5′)-P(4)- (2′-deoxycytidine 5′)tetraphosphate, tetrasodium salt],

a next-generation P2Y(2) receptor agonist for the treatment of cystic

ﬁbrosis. J. Pharmacol. Exp. Ther. 2002, 302, 871–880.

1

NMR (D₂O); 65:35 by HPLC). 64: H NMR (300 MHz, D₂O) δ

3.88 (m, 2H), 4.47 (d, 1H), 5.07 (m, 1H), 5.28 (m, 1H), 5.63 (d, J

) 6.6 Hz, 0.33H), 5.81 (d, J )6.3 Hz, 0.66H), 6.19 (m, 2H), 6.78

(dd, 1H), 7.01 (t, 1H), 7.26 (m, 9H), 8.44 (s, 0.33H), 8.46 (s, 0.66H),

8.49 (s, 0.66H), 8.59 (s, 0.33H). ³¹P NMR (121.47 MHz, D₂O) δ

4.86 (s, 0.35P), 4.89 (s, 0.65P). MS (ES): m/z 579.3 (M - H).

HPLC retention time (C₁₈, ion pairing): 13.62 min (65%), 14.19

(35%).

(7) Hechler, B.; Lenain, N.; Marchese, P.; Vial, C.; Heim, V.; Freund,

M.; Cazenave, J. P.; Cattaneo, M.; Ruggeri, Z. M.; Evans, R.; Gachet,

C. A role of the fast ATP-gated P2X₁cation channel in thrombosis of

small arteries in vivo. J. Exp. Med. 2003, 198, 661–667.

(8) Cattaneo, M.; Gachet, C. ADP receptors and clinical bleeding disorders.

Arterioscler., Thromb., Vasc. Biol. 1999, 19, 2281–2285.

(9) Hollopeter, G.; Jantzen, H. M.; Vincent, D.; Li, G.; England, L.;

Ramakrishnan, V.; Yan, R. B.; Nurden, A.; Julius, D.; Conley, P. B.

Identiﬁcation of the platelet ADP receptor targeted by antithrombotic

drugs. Nature 2001, 409, 202–207.

(10) Gachet, C. Identiﬁcation, characterization, and inhibition of the platelet

ADP receptors. Int. J. Hematol. 2001, 74, 375–381.

(11) Dogne, J. M.; de Leval, X.; Benoit, P.; Delarge, J.; Masereel, B.; David,

J. L. Recent advances in antiplatelet agents. Curr. Med. Chem. 2002,

9, 577–589.

((3aS,4R,6R,6aS)-6-(6-(3-Hexylureido)-9H-purin-9-yl)-2-(phe-

nylethynyl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihy-

drogen Phosphate (65). Compound 65 was prepared from 55

according to the method for the synthesis of 57. n-Hexylisocyanate

(4 equiv) was added in a single portion and the reaction heated at

80–85 °C overnight. Following alkaline treatment, workup (aqueous

NaHCO₃/ether), and preparative HPLC, the yield of 65 was 22%.

The product consisted of an approximately 48:52 ratio of acetal

1

diastereomers (45:55 by NMR (MeOD); 48:52 by HPLC). 65: H

(12) Ingall, A. H.; Dixon, J.; Bailey, A.; Coombs, M. E.; Cox, D.; McInally,

J. I.; Hunt, S. F.; Kindon, N. D.; Teobald, B. J.; Willis, P. A.;

Humphries, R. G.; Leff, P.; Clegg, J. A.; Smith, J. A.; Tomlinson, W.

Antagonists of the platelet P2T receptor: a novel approach to

antithrombotic therapy. J. Med. Chem. 1999, 42, 213–220.

(13) Leon, C.; Hechler, B.; Vial, C.; Leray, C.; Cazenave, J. P.; Gachet,

C. The P2Y₁receptor is an ADP receptor antagonized by ATP and

expressed in platelets and megakaryoblastic cells. FEBS Lett. 1997,

403, 26–30.

NMR (300 MHz, MeOD) δ 0.92 (t, 3H),1.37 (m, 6H), 1.64 (m,

2H), 3.38 (t, 2H), 4.07 (m, 2H), 4.60 (m, 0.45H),4.76 (m, 0.55H),

5.22 (dd, 0.55H), 5.34 (dd, 0.45H), 5.46 (dd, 0.55H), 5.56 (dd,

0.45H), 6.07 (s, 0.55H), 6.32 (s, 0.45H), 6.35 (d, J ) 3.3 Hz,

0.45H), 6.65 (d, J ) 3.3 Hz, 0.55H), 7.40 (m, 4H), 7.56 (m, 1H),

8.56 (d, 1H), 8.76 (s, 0.45H), 8.79 (s, 0.55H). ³¹P NMR (121.47

MHz, D₂O) δ 1.99 (s, 0.51P), 2.05 (s, 0.49P). MS (ES): m/z 585.4

(M - H). HPLC retention times (C₁₈, ion pairing): 14.80 min (48%),

15.97 (52%).

(14) Kauffenstein, G.; Hechler, B.; Cazenave, J. P.; Gachet, C. Adenine

triphosphate nucleotides are antagonists at the P2Y receptor. Thromb.

Haemost. 2004, 2, 1980–1988.

((3aS,4R,6R,6aS)-2-(Phenylethynyl)-6-(6-(3-phenylureido)-9H-

purin-9-yl)tetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methyl-dihydro-

gen Phosphate (66). Compound 66 was prepared from 55 according

to the method for the synthesis of 56, using phenyl isocyanate (5

equiv). The reaction was heated overnight at 30–35 °C. Following

workup and preparative HPLC, the yield of the title compound was

61%. The product consisted of a 50:50 ratio of acetal diastereomers

(47:53 by NMR (MeOD); 49:51 by HPLC). 66: ¹H NMR (300

MHz, MeOD) δ 4.05 (m, 2H), 4.64 (s, 0.47H), 4.76 (s, 0.53H),

5.25 (d, 0.52H), 5.39 (d, 0.48H), 5.48 (m, 0.52H), 5.58 (m, 0.48H),

6.04 (s, 0.53H), 6.33 (s, 0.47H), 6.39 (d, J ) 3.3 Hz, 0.49H), 6.70

(d, J ) 3.3 Hz, 0.51H), 7.11 (t, 1H), 7.40 (m, 6H), 7.59 (m, 3H),

8.67 (s, 1H), 9.02 (s, 0.47H), 9.05 (s, 0.53H). ³¹P NMR (121.47

MHz, D₂O) δ 5.07 (0.5P), 5.08 (0.5P). MS (ES): m/z 577.3 (M -

H). HPLC retention times (C₁₈, ion pairing): 13.59 min (49%), 14.42

(51%).

(15) Humphries, R. G. Pharmacology of AR-C69931MX and related

compounds: from pharmacological tools to clinical trials. Haemato-

logica 2000, 85, 66–72.

(16) Greenbaum, A. B.; Grines, C. L.; Bittl, J. A.; Becker, R. C.; Kereiakes,

D. J.; Gilchrist, I. C.; Clegg, J.; Stankowski, J. E.; Grogan, D. R.;

Harrington, R. A.; Emanuelsson, H.; Weaver, W. D. Initial experience

with an intravenous P2Y12 platelet receptor antagonist in patients

undergoing percutaneous coronary intervention: result from a 2-part,

phase II, multicenter, randomized, placebo- and active-controlled trial.

Am. Heart J. 2006, 151, 689e1–689e10.

(17) Husted, S.; Emanuelsson, H.; Heptinstall, S.; Clark, S.; Peters, G.

Greater and Less Variable Inhibition of Platelet Aggregation (IPA)

with AZD6140, the First Oral Reversible ADP Receptor Antagonist,

Compared with Clopidogrel in Patients with Atherosclerosis (The

DISPERSE Study). Presented at European Society of Cardiology

(ESC), Stockholm, Sweden, 7th September, 2005.

(18) (a) Scarborough, R. M.; Laibelman, A. M.; Clizbe, L. A.; Fretto, L. J.;

Conley, P. B.; Reynolds, E. E.; Sedlock, D. M.; Jantzen, H. Novel

tricyclic benzothiazolo[2,3-c]thiadiazine antagonists of the platelet

ADP receptor (P2Y(12)). Bioorg. Med. Chem. Lett. 2001, 11, 1805–

1808. (b) Scarborough, R. W.; Grant,C. M.; Zhang, X.; Cannon, H.;

Huang, W.; Mehrotra, M. 4-(6-Halo-7-substituted-2,4-dioxo-1,4-

dihydro-2H-quinazolin-3-yl)-phenyl-5-chloro-thiophen-2-yl-sulfony-

Acknowledgment. We thank Aris Ragouzeos for performing

the NOE experiments described in this article and for helpful

discussions surrounding the data. We would also like to

1024 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4

Douglass et al.

lureas and Forms and Methods Related Thereto. PCT Intl. Pat. Appl.

2007/056219 May 18, 2007.

N-ethyluronamide as A₁and A₃adenosine receptor agonists. J. Med.

Chem. 1998, 41, 3174–3185.

(39) Knorre, D. G.; Kurbatov, V. A.; Samukov, V. V. General method for

the synthesis of ATP gamma-derivatives. FEBS Lett. 1976, 70, 105–

108.

(40) Wakuda, H.; Ide, M.; Maeda, Y.; Kimura, J. Conformational Analysis

of 2′,3′-O-Alkylideneadenosines. Nippon Kagaku Kaishi 1998, 1, 40–

44.

(41) Bakthavachalam, V.; Lee-Gin, L.; Cherian, X. M.; Czarnik, A. W.

Synthesis, stereochemistry, intramolecular cyclization, and rates of

hydrolysis of adenosine 2′,3′-acetals. Carbohydr. Res. 1987, 170, 124–

135.

(19) Wang, Y. X.; Vincelette, J.; da Cunha, V.; Martin-McNulty, B.;

Mallari, C.; Fitch, R. M.; Alexander, S.; Islam, I.; Buckman, B. O.;

Yuan, S.; Post, J. M.; Subramanyam, B.; Vergona, R.; Sullivan, M. E.;

Dole, W. P.; Morser, J.; Bryant, J. A novel P2Y₁₂adenosine

diphosphate receptor antagonist that inhibits platelet aggregation and

thrombus formation in rat and dog models. Thromb. Haemost. 2007,

97, 847–55.

(20) Harker, L. A.; Boissel, J. P.; Pilgrim, A. J.; Gent, M. Comparative

safety and tolerability of clopidogrel and aspirin: results from CAPRIE.

CAPRIE Steering Committee and Investigators. Clopidogrel versus

aspirin in patients at risk of ischaemic events. Drug Saf. 1999, 21,

325–335.

(42) Further proof of structure for compounds 21, 31, 38, 45, 58, and 62

may be found in the Supporting Information.

(43) One plausible explanation to describe the selectivity of the acetal ring-

forming reactions can be illustrated by examining the following

transition states (A-D). The ﬁnal ring closure could be described as

a 5-endo trig cyclization of a free 2′ or 3′ hydroxyl group of the ribose

ring on to the oxonium ion derived from the complementary hydroxyl

group. The transition states are drawn using the 2′-hydroxyl as a

nucleophile and the 3′-oxonium ion as an electrophile for illustrative

purposes only. To minimize steric congestion, the conformation of

the ribose ring places the base and the side chain in pseudo equatorial

positions. Transition states A and B, which place the pivotal forming

stereogenic center of the new ring in a pseudo exo-orientation relative

to the ribose ring, should govern formation of kinetically favored

products. Transition states C and D, which place the pivotal forming

stereogenic center of the new ring in a pseudo endo-orientation, should

be less favored. Therefore, the kinetically favored product would be

derived from transition state A, which places the substituent in a pseudo

equatorial position. Kinetic products (products derived from aldehydes

that generate very unstable oxonium ions) should give predominantly

the cis-product. In circumstances where the formation of products is

under thermodynamic control, acetals derived from aldehydes that can

generate stable oxonium ions (cinnamaldehyde, benzaldehyde, and

ethynyl phenyl aldehyde), the favored product (whether derived from

an open or closed oxonium ion intermediate), would result from the

substituent being in the energetically more favored exo-face of the

5,5-bicyclic acetal. Thermodynamic products should give predomi-

nantly the trans-product.

(21) Gent, M.; Blakely, J. A.; Easton, J. D.; Ellis, D. J.; Hachinski, V. C.;

Harbison, J. W.; Panak, E.; Roberts, R. S.; Sicurella, J.; Turpie, A. G.

The Canadian American Ticlopidine Study (CATS) in thromboembolic

stroke. Lancet 1989, 1, 1215–1220.

(22) Bertrand, M. E.; Rupprecht, H. J.; Urban, P.; Gershlick, A. H. Double-

blind study of the safety of clopidogrel with and without a loading

dose in combination with aspirin compared with ticlopidine in

combination with aspirin after coronary stenting: the clopidogrel aspirin

stent international cooperative study (CLASSICS). Circulation 2000,

102, 624–629.

(23) Savi, P.; Pereillo, J. M.; Uzabiaga, M. F.; Combalbert, J.; Picard, C.;

Maffrand, J. P.; Pascal, M.; Herbert, J. M. Identiﬁcation and biological

activity of the active metabolite of clopidogrel. Thromb. Haemost.

2000, 84, 891–896.

(24) Nguyen, T. A.; Diodati, J. G.; Pharand, C. Resistance to clopidogrel:

a review of the evidence. J. Am. Coll. Cardiol. 2005, 45, 1157–

1164.

(25) Bennett, C. L.; Connors, J. M.; Carwile, J. M.; Moake, J. L.; Bell,

W. R.; Tarantolo, S. R.; McCarthy, L. J.; Sarode, R.; Hatﬁeld, A. J.;

Feldman, M. D.; Davidson, C. J.; Tsai, H. M. Thrombotic thromb-

ocytopenic purpura associated with clopidogrel. N. Engl. J. Med. 2000,

342, 1773–1777.

(26) Weerakkody, G. J.; Jakubowski, J. A.; Brandt, J. T.; Payne, C. D.;

Naganuma, H.; Winters, K. J. Greater inhibition of platelet aggregation

and reduced response variability with prasugrel versus clopidogrel:

an integrated analysis. J. CardioVasc. Pharmacol. Ther. 2007, 3, 205–

12.

(27) Tantry, U. S.; Bliden, K. P.; Gurbel, P. A. Prasugrel. Expert Opin.

InVest. Drugs 2006, 15, 1627–33.

(28) Zamecnik, P. C.; Stephenson, M. L.; Janeway, C. M.; Randerath, K.

Enzymatic synthesis of diadenosine tetraphosphate and diadenosine

triphosphat with a puriﬁed lysyl-sRNA synthetase. Biophys. Res.

Commun. 1966, 24, 91–97.

(29) Pintor, J.; Rotllan, P.; Torres, M.; Miras-Portugal, M. T. Characteriza-

tion and quantiﬁcation of diadenosine hexaphosphate in chromafﬁn

cells: granular storage and secretagogue-induced release. Anal. Bio-

chem. 1992, 200, 296–300.

(30) Rotllan, P.; Miras-Portugal, M. T. Adenosine kinase from bovine

adrenal medulla. Eur. J. Biochem. 1985, 151, 365–71.

(31) Hilderman, R. H.; Christensen, E. F. P1,P4-diadenosine 5′ tetraphos-

phate induces nitric oxide release from bovine aortic endothelial cells.

FEBS Lett. 1998, 427, 320–324.

(32) Pintor, J.; Diaz-Hernandez, M.; Gualix, J.; Gomez-Villafuertes, R.;

Hernando, F.; Miras-Portugal, M. T. Diadenosine polyphosphate

receptors. from rat and guinea-pig brain to human nervous system.

Pharmacol. Ther. 2000, 87, 103–115.

(33) Harrison, M. J.; Brossmer, R. Inhibition of platelet aggregation and

the platelet release reaction by alpha, omega diadenosine polyphos-

phates. FEBS Lett. 1975, 54, 57–60.

(34) Rapaport, E.; Zamecnik, P. C. Presence of diadenosine 5′,5′′′-P1, P4-

tetraphosphate (Ap4A) in mamalian cells in levels varying widely with

proliferative activity of the tissue: a possible positive “pleiotypic

activator”. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 3984–3988.

(35) Chan, S. W.; Gallo, S. J.; Kim, B. K.; Guo, M. J.; Blackburn, G. M.;

Zamecnik, P. C. P1,P4-dithio-P2,P3-monochloromethylene diadenosine

5′,5′′′-P1,P4-tetraphosphate: a novel antiplatelet agent. Proc. Natl.

Acad. Sci. U.S.A. 1997, 94, 4034–4039.

(36) Bhattacharyya, S.; Kerzmann, A.; Feig, A. L. Fluorescent analogs of

UDP-glucose and their use in characterizing substrate binding by toxin

A from Clostridium difﬁcile. Eur. J. Biochem. 2002, 269, 3425–3432.

(37) Martin, P. L.; Gero, T. W.; Potts, A. A.; Cusack, N. J. Structure-

activity studies of analogs of ꢀ, γ-methylene-ATP at P2x-purinore-

ceptors in the rabbit ear central artery. Drug DeV. Res. 1995, 36, 153–

165.

(44) Shaver, S. R.; Rideout, J. L.; Pendergas, W.; Douglass, J. G.; Brown,

E. G.; Boyer, J. L.; Patel, R. I.; Redick, C. C.; Jones, A. C.; Picher,

M.; Yerxa, B. R. Structure-activity relationships of dinucleotides:

Potent and selective agonists of P2Y receptors. Purinergic Signal 2005,

1, 183–191.

(45) Hiratsuka, T. New ribose-modiﬁed ﬂuorescent analogs of adenine and

guanine nucleotides available as substrates for various enzymes.

Biochim. Biophys. Acta 1983, 742, 496–508.

(46) Wright, M.; Miller, A. D. Synthesis of novel ﬂuorescent-labeled

dinucleoside polyphosphates. Bioorg. Med. Chem. Lett. 2004, 14,

2813–2816.

(47) In addition to the aforementioned regiochemistry problems, the ribose

esters were also less stable as solids and in solution than acetals. At

room temperature and a pH of greater than 7, which would be the

typical physiological situation, compounds 15–17 showed signiﬁcant

decomposition after a few days. The acetal compounds were stable in

solution under these conditions. Solid ester compounds typically were

stored at -80 °C in their ammonium salt forms to avoid decomposition

over the long term, whereas solid acetals (in their sodium or

ammonium salt forms) could be stored on the benchtop for many

months without ill effect. The general trend for stability of the aromatic

acetals under acidic conditions was styryl < phenyl < benzyl.

(48) In general, the rank order of hydrolytic stability of the phosphates

that we observed was dinucleoside polyphosphates > nucleoside

monophosphates > nucleoside diphosphates > nucleoside triphos-

phates. Dinucleoside polyphosphates and nucleoside monophosphates

(38) Baraldi, P. G.; Cacciari, B.; Pineda de Las Infantas, M. J.; Romagnoli,

R.; Spalluto, G.; Volpini, R.; Costanzi, S.; Vittori, S.; Cristalli, G.;

Melman, N.; Park, K. S.; Ji, X.; Jacobson, K. A. Synthesis and

biological activity of a new series of N⁶-arylcarbamoyl, 2-(Ar)alkynyl-

N⁶-arylcarbamoyl, and N⁶-carboxamido derivatives of adenosine 5′-

Modiﬁed Nucleotides as P2Y₁₂Receptor Antagonists

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 1025

modiﬁed with the lipophilic moieties described in this article could

be stored in their salt forms (sodium, ammonium, or tributylammo-

nium) at ambient temperature for months without signiﬁcant decom-

position. As noted in the preparation of 41, it was nearly impossible

to avoid some degradation of the nucleoside diphosphate to the

monophosphate (56), even with optimal handling. The corresponding

triphosphate analogue to 41and56(not included in this article) proved

to be much more troublesome and was not pursued.

Adenine Nucleotides as Reversible Inhibitors of P2Y₁₂-Mediated

Platelet Aggregation. 4th International Symposium of Nucleosides and

Nucleotides, Chapel Hill, North Carolina, June, 2004.

(50) Arunlakshana, O.; Schild, H. O. Some quantitative uses of drug

antagonists. Br. J. Pharmacol. Chemother. 1959, 1, 48–58.

(51) Johnson, F. L.; Boyer, J. L.; Leese, P. T.; Crean, C.; Krishnamoorthy,

R.; Durham, T.; Fox, A. W.; Kellerman, D. J. Rapid and reversible

modulation of platelet function in man by a novel P2Y(12) ADP-

receptor antagonist, INS50589. Platelets 2007, 18, 346–356.

(49) These data have been partially reported in poster format at the 4th

International Symposium of Nucleosides and Nucleotides, Chapel Hill,

North Carolina, June, 2004. Douglass, J. G.; Patel, R. I.; Redick, C. C.;

Jones, A. C.; Shave, S. R.; Boyer, J. L. (Poster 70W) Lipophilic

JM701348D

Article Doi

Lipophilic modifications to dinucleoside polyphosphates and nucleotides that confer antagonist properties at the platelet P2Y12 receptor

DOI: 10.1021/jm701348d

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm701348d

Full text of DOI:10.1021/jm701348d

Products guided by the article

R&D Labs maybe for 1012303-33-1

Relevant to this article

Hot Product