Inhibition of ADP-triggered blood platelet aggregation by diadenosine polyphosphate analogues

Article abstract of DOI:10.1016/S0960-894X(02)00318-9

The synthesis and biological evaluation of new diadenosine polyphosphate analogues on blood platelet aggregation are reported. The most active are compounds with a sulfur atom replacing one or both non-bridging oxygens at phosphorus bound to adenosyl resi

Full text of DOI:10.1016/S0960-894X(02)00318-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1959–1962
Inhibition of ADP-Triggered Blood Platelet Aggregation by
Diadenosine Polyphosphate Analogues
Bogdan Walkowiak,a,b,* Janina Baraniak,c Czeslaw S. Cierniewskia and Wojciech Stecc
aDepartment of Molecular and Medical Biophysics, Institute of Physiology and Biochemistry, Medical University of Lodz,
Lodz, Poland
bDepartment of Biophysics, Institute of Material Engineering, Technical University of Lodz, Lodz, Poland
cDepartment of Bioorganic Chemistry, Center for Molecular and Macromolecular Research, Polish Academy of Science,
Lodz, Poland
Received 27 December 2001; revised 23 April 2002; accepted 2 May 2002
Abstract—The synthesis and biological evaluation of new diadenosine polyphosphate analogues on blood platelet aggregation are
reported. The most active are compounds with a sulfur atom replacing one or both non-bridging oxygens at phosphorus bound to
adenosyl residues and hydroxymethyl groups of bis(hydroxymethyl)phosphinic acid. # 2002 Elsevier Science Ltd. All rights
reserved.
Blood platelets play an important role in prevention of
blood circulation frommassive bleeding caused by vas-
cular damage, but on the other hand, they are also
responsible for intravascular plaque formation resulting
in arteriosclerosis. Thus balanced platelet reactivity is
crucial for proper circulation. It is well documented that
insufficiently reactive platelets are responsible for
bleeding tendency in uremia,1 although platelet hyper-
activity increases the risk of arteriosclerosis and cor-
onary disease.2 Platelets can be activated by several
physiological agonists like adenosine diphosphate
(ADP), thrombin, platelet activating factor (PAF), col-
lagen, and others,3 but the most important seems to be
ADP, which can evoke platelet activation and aggrega-
tion, and is responsible for proper formation of the
second wave of platelet aggregation, followed by ADP
for ADP action,8 although both are highly unstable in
circulation.
Activation of platelets by ADP is mediated by several
purinoceptors showing distinct specificity.9 For exam-
ple, P2Y and P2T receptors are antagonized by ATP and
AMP, whereas P2X is agonized by ATP but not by
ADP.10 P2X forms an ionotropic channel which allows
sodium and calcium ions to pass the platelet membrane.
Though it can trigger platelet activation accompanied by
shape change, aggregation of platelets requires involve-
ment of P2Y and P2T receptors as well.11 Diadenosine
tetraphosphates (Ap4A), like other diadenosine poly-
phosphates, are P2X purinoceptor agonists,12 which can
evoke platelet activation, but at the same time these
agents act as inhibitors for ADP action on P2Y and P2T
receptors, leading to inhibition of platelet aggregation.
Complete and actual information about the meaning of
above mentioned receptors can be found in a review
article.13
4
release fromplatelet granules. Attempts for treatment
of overactive platelets are numerous, but most fre-
quently aspirin is used for inhibition of platelet cyclo-
oxygenase activity and thromboxanes formation.5 Also
disintegrins and RGD-analogues are used for inhibition
of fibrinogen binding,6 but clinically useful for inhibi-
tion of ADP binding to its receptor are ticlopidine and
clopidogrel.7 It is well known that both ATP and AMP
nucleotides are very potent competitors and antagonists
Recently, a number of stable analogues of adenosine
nucleotides with structure related to diadenosine poly-
phosphates were described in terms of their antiplatelet
activity.14,15 Among them, diadenosine 50,50000-P1,P4-
tetraphosphate (Ap4A) and its chloromethylene thio-
phosphate analogue ApspCHClppsA appeared to be
strong antithrombotic agents that block a very early stage
of thrombus formation. Further to these observations, in
*Corresponding author. Tel.: +48-42-678-9433; e-mail: bogdan.
walkowiak@csk.am.lodz.pl
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00318-9

1960
B. Walkowiak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1959–1962
this report we describe new analogues of ApnA with
antiplatelet activity and focus on their structure–activity
relationships. These compounds, synthesized based upon
the oxathiaphospholane methodology,16 show several
common structural features. First of all, adenosine 50-O-
phosphorothioate units are linked together through
polyols, diamines, dimercaptoalkanes or bis(hydroxy-
methyl)phosphinic acid residues.17 Interestingly, some
of them, like 1,3-bis(adenosine-50-O-phosphorothioyl)-
glycerol, have appeared to be superior inhibitors of
FHIT protein.18 Data presented in this report show that
these compounds, particularly isosteric analogues of
Ap4A, are also strong inhibitors of ADP-induced plate-
prepared by reaction of O1,O3-(2-thiono-1,3,2-oxa-
thiaphospholanyl)-bis(hydroxymethyl)phosphinic acid
O-methyl0 ester with one molar equivalent of
0
N6,N6,O2 ,O3 -tetrabenzoyladenosine in the presence of
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), followed by
molar equivalent of b-cyanoethanol.0 Reaction of two
0
molar equivalents of N6,N6,O2 ,O3 -tetrabenzoyl-
adenosine
50-O-(2-thiono-1,3,2-oxathiaphospholane)
with O,O-dimethyl-d,l-tartrate and in the presence of
DBU in acetonitrile solution furnished compound 6.
Compound 12 has also been obtained in a two step
procedure: treatment of 1,3-diaminopropane with two
molar equivalents of 2-chloro-1,3,2-oxathiaphospholane
in the presence of triethylamine and elemental sulfur
0
0
let aggregation. N6,N6,O2 ,O3 -Tetrabenzoyladenosine
50-O-(2-thiono-1,3,2-oxathiaphospholane)17 was used to
synthesize analogues with chemical formulas presented
in Figures 1 and 2.
gave
N1,N3-bis(1,3,2-oxathiaphospholanyl)diamino-
propane in 72% yield [31P NMR (CDCl3) 96.7 ppm;
FAB-MS (Mꢀ1) m/z 349]. This compound i0n reaction
0
with two molar equivalents of N6,N6,O2 ,O3 -tetra-
The synthesis of compounds 3–5, 7–10 and 13 was
described recently.17 For the synthesis of compounds 1
and 11, the following procedure has been used: pyridine
benzoyladenosine in acetonitrile solution, in the pres-
ence of DBU, provided desired compound 12. All
compounds were purified by means of DEAE Sephadex
A-25 ion exchange chromatography, using a linear gra-
dient of ammonium bicarbonate buffer (pH 7.5) as an
eluent. Their structure and physicochemical character-
istics were verified by MS-MALDI and 31P NMR
(Table 1).
0
0
solution of N6,N6,O2 ,O3 -tetrabenzoyladenosine con-
taining 3-fold molar excess of elemental sulfur was
treated with two molar equivalents of 2-chloro-1,3,2-
0
0
dithiaphospholane. Resulting N6,N6,O2 ,O3 -tetraben-
zoyladenosine 50-O-(2-thiono-1,3,2-oxathiaphospholane)
[isolated in 68% yield, 31P NMR (CDCl3) 124.3 ppm;
FAB-MS (Mꢀ1) m/z 836.4] in DBU-assisted reaction
with half-molar equivalent of bis(hydroxymethyl)phos-
phinic acid O-methyl ester, in acetonitrile solution,
provided compound 1 (Scheme 1).
Aggregation test was performed with the recently pub-
lished method involving microplate reader.19 Blood was
drawn fromhealthy human volunteers who were not
medicated by any drug during at least 2 weeks prior to
the examination. Informed consent was obtained from
each donor and the study was performed according to
the Helsinki Declaration. Blood was anticoagulated
with 3.8% sodiumcitrate added in 1:9 ratio. Freshly
collected human citrated blood was subjected to cen-
trifugation (200g, 10 min, rt), and platelet-rich plasma
0
0
Similarly, N6,N6,O2 ,O3 -tetrabenzoyladenosine 50-O-(2-
thiono-1,3,2-dithiaphospholane) in reaction with gly-
cerol provided compound 11. Compound
2 was
Figure 1.
Figure 2.
Scheme 1.

B. Walkowiak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1959–1962
1961
(PRP) was used for the aggregation test. Platelet-poor
plasma (PPP) was obtained by subsequent centrifuga-
tion of PRP at room temperature for 20 min at 1000g.
The instrument used for aggregation test (EL340 Bio-
Tek Instruments) could maintain temperature 37 ꢁC and
continuously shake (forth and back) the microplate
between readings (20 s). Readings were performed at
595 nmand repeated every 30 s. The results of reading
were stored in computer memory for further evaluation.
Flat-bottom96-well Costar microplates were used. The
studied analogues and AMP (Sigma Aldrich), dissolved
in phosphate buffer (pH 7.4), were prepared as series of
dilutions in a microplate (20 mL of each), and to each
sample dilution 100 mL of PRP were dispensed. Follow-
ing tender shaking, samples were preincubated for
30 min at rt, and then 20 mL of ADP (final concentration
5 mM) were injected into each well, shaken and obser-
vations of aggregation process were started. For a con-
trol, some wells were processed in the presence of 20 mL
of phosphate buffer instead of the studied sample,
whereas others were filled with PPP alone.
In this study, 13 analogues of Ap4A with chemical
structures given in Figures 1 and 2 were synthesized
as described above and tested in terms of their inhibi-
tory effects on ADP-induced platelet aggregation.
Compounds 2–5, 8–10, 12 and 13 were studied as unsep-
arated mixtures of P-diastereomers. In parallel
adenosine 50-monophosphate (AMP) was used as a
positive control. All compounds were used as sodium
salts.
We have found three analogues showing very strong
potency for inhibition of blood platelet aggregation
evoked in PRP by ADP (Fig. 3), comparable with that
elicited by AMP. Assigned IC50 values for each
compound are given in Table 1.
Table 1. The physicochemical characteristics of compounds 1–13 and
their IC50 values
Compd
31P NMR
(ppm)
MALDI-MS
(Mꢀ1) m/z
IC50
(mM)
In our tests compounds 1–3 were able to inhibit by 50%
aggregation process if used at 10–15 mM concentration.
In the same experiments 50% inhibition was observed
for 7 mM AMP. The next four compounds 4–7 exhibited
medium inhibitory potency with 50% inhibition
observed for concentrations ranging from40 to 80 mM.
The last group consisted of six compounds 8–13 with
low or no observable inhibitory potency in our test.
Compounds 1, 3 and 4 can be considered as isosteric
analogues of Ap4A; adenosyl moieties are separated by
seven atoms, albeit they are not isoelectric, since on the
contrary with Ap4A they possess three formal negative
charges. An important feature of their structure is the
lack of pyrophosphoryl linkages, an important element
of their chemical stabilization and resistance to nucleo-
lytic enzymes, that potentially promotes them as orally
administered drugs. Conservation of the distance
between adenosyl moieties and three negative charges
is salutary for occupation of active site of biological
entity they interact with, since compounds 5 and 6 with
eight atomspacers are less active. Some years ago it has
been reported that adenosine-50-phosphorothioate also
inhibited platelet aggregation.20 It should be empha-
sized that key structural motif in the most active com-
pounds is the sulfur atomreplacing one or both non-
bridging oxygens at phosphorus bound to adenosyl
residues and hydroxymethyl groups of bis(hydroxy-
methyl)phosphinic acid. Our best inhibitor, compound
1, contains two phosphorodithioate functions within a
seven atomscaffold, which additionally stabilize that
compound against nucleolytic enzymes. The high avid-
ity of phosphorodithioates incorporated in oligonucleotide
constructs towards protein has been reported.21
1
2
3
4
5
6
7
8
116.7; 30.7
59.5; 44.9
58.5; 30.3
58.3; 30.3; 2.1
57.1
58.8; 57.7
30.5; 1.7
56.8
56.6; 45.9
56.3
115.9
60.2; 58.5
57.6
847
566
815
799
812
840
783
782
877
1516
814
1125
1127
10
13
16
40
70
75
80
140
>160
>160
—
—
—
9
10
11
12
13
That consistent picture suggesting interaction of our
compounds with P2T receptor is not perturbed by com-
pound 2, bearing only one adenosyl residue attached
to a seven atomscaffold P(X,Y ꢀ)OCH2P(O,Oꢀ)CH2O
P(X,Yꢀ) (where the X and Y are both sulfur atoms, or
sulfur and oxygen, respectively) and demonstrating high
inhibitory effect to platelet aggregation. P2T purinoceptor,
Figure 3. The effect of Ap4A analogues on platelet aggregation trig-
gered by ADP. Increasing concentrations of the analogues were added
to platelet-rich plasma and preincubated for 30 min at room tempera-
ture in flat-bottom96-well Costar microplates. Then, 20 mL of ADP
(final concentration 5 mM) were injected into each well, shaken and
observation of aggregation process, using EL340 BioTek Instruments,
was continued for 10 min with readings performed at 595 nm every
30 s.

1962
B. Walkowiak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1959–1962
serving as binding site for ADP, requires the presence of
only one adenosyl residue for binding and recognition,
and therefore can be competitively occupied by a com-
pound like 2, albeit that highly speculative explanation
requires further studies. In summary, in this communi-
cation we present a novel class of compounds with high
potency of inhibition of platelet aggregation process.
They are easily available due to methodology elabo-
rated in this Laboratory, and of highest activity com-
pound 1, as a P-achiral species, does not require
separation into diastereomers.
7. Topol, E. J.; Easton, J. D.; Amarenco, P.; Califf, R.; Har-
rington, R.; Graffagnino, C.; Davis, S.; Diener, H. C.;
Ferguson, J.; Fitzgerald, D.; Shuaib, A.; Koudstaal, P. J.;
Theroux, P.; Van-de-Werf, F.; Willerson, J. T.; Chan, R.;
Samuels, R.; Ilson, B.; Granett, J. Am. Heart J. 2000, 139, 927.
8. Gachet, C.; Cazenave, J. P. Nouv. Rev. Fr. Hematol. 1991,
33, 347.
9. Kunapuli, S. P. Trends Pharmacol. Sci. 1998, 19, 391.
10. Mahaut-Smith, M. P.; Eninion, S. J.; Rolf, M. G.; Evans,
R. J. Br. J. Pharmacol. 2000, 131, 108.
11. Rolf, G. M.; Brearley, C. A.; Mahaut-Smith, M. P.
Thromb. Haemost. 2001, 85, 303.
12. Sage, S. O.; MacKenzie, A. B.; Jenner, S.; Mahaut-Smith,
M. P. Prost. Leukot. Essent. Fatty Acids 1997, 57, 435.
13. Gachet, C. Thromb. Haemost. 2001, 86, 222.
14. Chan, S. W.; Gallo, S. J.; Kim, B. K.; Guo, M. J.; Black-
burn, G. M.; Zamecnik, P. C. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 4034.
15. Zamecnik, P. C.; Kim, B. K.; Guo, M. J.; Taylor, G.;
Blackburn, M. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 2370.
16. Guga, P.; Koziolkiewicz, M.; Okruszek, A.; Stec, W. J. In
Applied Antisense Oligonucleotide Technology; Stein, C., Krieg,
A., Eds.; John Wiley and Sons: New York, 1998; p. 23.
17. Baraniak, J.; Wasilewska, E.; Korczynski, D.; Stec, W. J.
Tetrahedron Lett. 1999, 40, 8603.
Acknowledgements
This project was financially assisted by the State Com-
mittee for Scientific Research (KBN), grants no. 4 P05F
006 17 and PBZ KBN 008/T09/98 (to W.J.S.).
References and Notes
1. Horowith, H. I.; Bronx, N. J. Arch. Int. Med. 1970, 12, 823.
2. Becker, R. C. Cardiology 1999, 79, 49.
18. Varnum, J. M.; Baraniak, J.; Kaczmarek, R.; Stec, W. J.;
Brenner, C. BMC Chemical Biology 2001, 1, 3 (www.bio-
medcentral.com/1472-6769/1/3).
19. Walkowiak, B.; Kesy, A.; Michalec, L. Thromb. Res. 1997,
87, 95.
20. Gough, G. R. J. Med. Chem. 1978, 21, 520.
21. Yang, X.-B.; Fennewald, S.; Luxon, B. A.; Aronson, J.;
Herzog, N. K.; Gorenstein, D. G. Bioorg. Med. Chem. Lett.
1999, 9, 3357.
3. Holmsen, H. J. Clin. Invest. Eur. 1994, 24, 3.
4. Lages, B.; Weiss, H. J. Thromb. Haemost. 1980, 43, 147.
5. McAdam, B.; Keimowitz, R. M.; Maher, M.; Fitzgerald,
D. J. J. Pharmacol. Exp. Ther. 1996, 277, 559.
6. McLane, M. A.; Gabbeta, J.; Rao, A. K.; Beviglia, L.;
Lazarus, R. A.; Niewiarowski, S. Thromb. Haemost. 1995, 74,
1316.