From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

Prostacyclin (Prostaglandin 12, PGI2) Inhibits

Platelet Adhesion and Thrombus
Formation on Subendothelium

By Harvey J. Weiss and Vincent T. Turitto

Prostaglandin I2 (prostacyclin, PGI2), a sub- adhesion and thrombus formation on suben-

stance synthesized in the wall of blood dothelium was inhibited in blood containing
vessels, has been previously shown to inhibit 10 nM PGI2. PGI2 appears to reduce adhesion
the aggregation of platelets in Stirred plate- by inhibiting platelet spreading. These find-
let-rich plasma. We used a method in which ings suggest that PGI2 could regulate the
segments of deendothelialized rabbit aorta deposition of platelets on vascular sur-
are perfused at arterial shear rates with faces.
human blood and found that both platelet

T HE PLATELETS in circulating blood do not adhere to each other or to the

intact endothelium of blood vessels. However, if the endothelium becomes
detached, the platelets can adhere and spread on the subendothelial surface, and
these adherent platelets can recruit others to produce a platelet thrombus.
Platelet-induced thrombosis, as well as thromboxane A2,2 serotonin, a mitogenic
factor,3 and other substances released by platelets,4 may mediate a variety of
pathologic processes.5 Hence an understanding of the mechanisms that regulate the
deposition of platelets on vascular surfaces is important.
Recent studies have shown that the intima6 and other layers of the vessel wall7
produce a potent inhibitor of platelet aggregation. The structure of this inhibitor,
prostacyclin or prostaglandin I, (PGI2), has been determined as 9-deoxy-6,9-
a-epoxy-5-PGF,, an intermediate in the transformation of prostaglandin endo-
peroxides into the stable end-product 6-keto-PGF,(,.8 The platelet-inhibitory
properties of PGI2 have been demonstrated by showing that extracts or microsomal
fractions of the vessel wall can inhibit arachidonic acid-induced platelet aggrega-
tion in platelet-rich plasma (PRP).67 Extracts of endothelial cells,9 skin fibro-
blasts,#{176}and arterial smooth muscle cells#{176}
grown in cell culture also inhibit platelet
aggregation and secretion. The above observations have led to the hypothesis that
PG!, may be a naturally occurring substance that prevents the formation of platelet
thrombi on injured blood vessels.67 Whether or not P012 could, in addition, prevent
platelet adhesion could not be determined with the test systems used.
Baumgartner has developed an ex vivo perfusion chamber for studying the
interaction of human platelets in flowing citrated whole blood with the subendo-
thelium of rabbit aorta at shear rates comparable to those in the arterial
circulation. With this technique, both platelet adhesion and thrombus formation

From the Department of Medicine (Division of Hematology). The Roosevelt Hospital and Columbia
University College of Physicians and Surgeons, New York, N.Y.
Submitted July 7, 1978; accepted October 3. 1978.
Supported in part by USPHS Grant ilL 14595 and a Grant-in-Aid from the American Heart
Association with funds contributed in part by the New York Heart Association.
Address for reprint requests: Harvey J. Weiss, M.D., The Roosevelt Hospital. 428 W. 59th St., New
York, N.Y. 10019.
(c /979 by Grune & Stratton, Inc. 0006-497//79/5302-0009$0/.00/0

244 Blood, Vol. 53, No. 2 (February), 1979

From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

PROSTACYCLIN INHIBITS PLATELET ADHESION 245

on the subendothelial surface can be observed and quantified, and recent studies by
Tschopp et al. showed that the deposition patterns of human platelets on rabbit and
human subendothelium are similar.2 The method has been used to identify various
defects of platelet adhesion and thrombus formation in patients with bleeding
disorders.34 For example, platelet thrombus formation is specifically decreased in
thrombasthenia,5 whereas adhesion is decreased in Bernard-Soulier syndrome6
and von Willebrands
We used this technique to examine the effects of PGI2 on platelet adhesion and
thrombus formation. The studies were performed on citrated human blood to which
varying concentrations of PGI2 had been added. We also prepared PRP and studied
platelet aggregation in the aggregometer.

MATERIALS AND METHODS

PG!2. a generous gift from Dr. K. C. Nicolaou, was synthesized from the methyl ester of
prostaglandin F,,, as previously described.7 A stock solution containing I mM PGI2 in a vehicle (I mM
Nal, 3 mM sodium ethoxide, and 95% EtOH, pH 1 1 .0) was stored at - 20#{176}C.Prior to use, vehicle and
PGI2 were diluted 1/10 with 0.15 M NaCI:0.15 M Iris (pH 9.0) (2:1, v/v). Further dilutions of the
PGI2 were in this Tris-saline buffer.
Blood. Venous blood from normal human subjects was mixed with 0. 1 vol 3.2% sodium citrate.8 For
the adhesion studies, the citrate concentration was adjusted to 19.7 mM in plasma.#{176} The blood was
incubated for 5 mm at 37#{176}C
with either nothing or I :99 (v/v) of the following: normal saline (NS), a
I : I 0 dilution of vehicle (V), or PG 12.
Platelet adhesion and thrombus formation. Everted segments of rabbit aorta. I 4 mm in length,
from which the endothelium had been removed by balloon catheter1#{176}14 were mounted on the inner core
of an annular perfusion chamber (either our standard chamberiiLI or a modified chamber with smaller
dimensions.#{176} Blood maintained at 37#{176}C
was circulated in a closed system through the chamber1#{176}14 at a
flow rate ofeither 160 mI/mm for 10 min#{176}(standard chamber, calculated wall shear rate of 800/sec1
or 40 mI/mm for 5 mm (small chamber, wall shear rate of 2600/sec. In one study, we also used
everted vessel segments previously digested with a-chymotrypsin,4 which produces a surface whose high
thrombogenicity is due to the exposure of fibrillar collagen.4 After perfusion, segments were removed
from the chamber for further fixation, processing, and embedding.iW4 Vessel sections 0.8 m in
thickness were evaluated morphometrically1314 around the entire circumferenceofthe vessel at i0zm
intervals as either contact (C, platelets contact subendothelium but do not spread upon it). spread (S,
platelets spread upon subendothelium), or thrombus (I, platelet thrombus 5 m or more in height
superimposed on spread platelets). Two sections approximately 3 and 10 mm from the proximal end of
the vessel segment were evaluated for each segment, and the values of C, 5, and I were averaged.
Platelet adhesion is defined as C + 5; lOOT/S is a measure of the extent to which thrombi have formed
on spread platelets and is an index of platelet-to-platelet interaction (aggregation).
Platelet aggregation. The effects of PGI2 on platelet-subendothelium interaction were compared at
various concentrations of the drug, with its effect on platelet aggregation studied by the more
conventional technique using the aggregometer. Blood was centrifuged at 20#{176}C
and I 500 g for I 35 sec to
obtain PRP. Platelet aggregation in the aggregometerit was induced with final concentrations of the
following agonists (added in 0.05 vol): 5 sM adenosine diphosphate (ADP) (Sigma Chemical, St. Louis,
Mo.); 5 zM epinephrine; 1.5 mg/mI ristocetin (Abbott Pharmaceutical, N. Chicago, Ill.); and two
dilutions of a collagen suspension5 (the dilute collagen suspension induces platelet aggregation that can
be inhibited by aspirin,it whereas aspirin does not inhibit aggregation by the more concentrated
suspension).

RESULTS

Platelet aggregation in the aggregometer. As seen in Fig. 1, 10 nM PGI2 (in

the blood from which the PRP was prepared) inhibited the change in platelet shape
that normally precedes aggregation induced by ADP or collagen. Shape change
was inhibited even (as in Fig. I D) when aggregation was not. Figure 2 depicts the
From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

246 WEISS AND TURITTO

A B P0)7

PGt

it
C #{248},,IllllmIPPIP P 012

Time

Fig. 1. Effect of PGI2 on platelet shape change and aggregation. Platelet aggregation in PAP prepared
from blood that contained either vehicle (V) or 10 nM PGI2. (A) 5 sM ADP; (B) 5 sM epinephrine; (C) dilute
collagen; (D) concentrated collagen. Note that PGI2 inhibited platelet shape change (SC) even where (as in
D) aggregation was not inhibited.

100

70

0 60

50

0.
40
j
30

20

10

I0 20 50 100 500 1000

PG2 concentratton (nM)

Fig. 2. Dose-response inhibition of platelet aggregation by PGI2. Platelet aggregation in PAP prepared
from blood that contained either normal saline (NS), vehicle (V), or PGI2. A, ADP (5 MM); , epinephrine
(5 MM); . dilute collagen;#{149},concentrated collagen; X, ristocetln (1.5 mg / ml).
From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

PROSTACYCLIN INHIBITS PLATELET ADHESION 247

A i_p

.-

_________ . --

:.;- -

Fig. 3. Platelet InteractIon with subendothellum. Vessel segments were exposed to flowing blood at 40
mImin for 5 mm (wall shear rate = 2600 sec ). (A) Blood Incubated with 0.01 vol vehIcle (diluted 1:10 in
Tris-sallne) for 5 mm before perfusion. (B) Same blood incubated with 0.01 vol PGI2 (final concentration 10
nM). Note elimination of platelet thrombi by PGI2. Platelet adhesion to subendothelium was partially
inhibited (see Table 1). Bar. 10 1m.

dose-response curves for PGI2 and platelet aggregation by various agonists.

Addition of PGI2 to blood in concentrations of 10-20 nM resulted in a 90 or
greater reduction in platelet aggregation induced by ADP or a dilute suspension of
collagen. Higher concentrations of PGI2 were needed to inhibit aggregation
induced by ristocetin, the more concentrated suspension of collagen, and epineph-
rine (which is usually considered to be a weak agonist for platelet aggregation).
Platelet interaction with subendotheliutn. We found that PGI2 strongly inhib-
ited the formation of platelet thrombi, as shown in the photomicrograph of Fig. 3
and in Tables I and 2. The most striking inhibition was observed in the studies with
the thrombogenic a-chymotrypsin-digested subendothelium (Table 2). Addition of
PGI2 to blood in a concentration of 10 nM resulted in a reduction of lOOT/S values
from 59.1% 8.7% (observed with vehicle-treated blood) to 6.1% 2.8%. With
100 nM PGI, this value was reduced further to 2.5% 1.1%, and thrombi were
eliminated completely with 1000 nM PG 12. Results of studies using (undigested)
subendothelium are shown in Table 1 and Fig. 3. PGI2 (10 nM) virtually
From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

248 WEISS AND TURITTO

Table 1. Platelet Interaction With Subendothelium

Shear Rate 800 sec 10 mm Shear Rat e 2600 sec 5 mm

C+S lOOT/S C+S lOOT/S

Addition (Adhesion( (%( 1 OOC/C + S (Aggregation( (Adhesion( (%( 1 OOC/C + S (Aggregation(

None 78.8 4.0 1.1 0.2 9.0 78 75.9 4.1 29 04 15.1 92

Vehicle 79.7 8.6 0.8 0.4 9.3 5.1 72.8 6.3 52 14 158 i 94
PGI7 lOnM 65.9 42INS( 3.5 1.1 1.0 0.5 40.2 7.8 231 68 02 02
PGI2 100nM 54.8 13.1(N5( 12.7 5.4 0.7 0.5 33.6 4.Ot 18.7 3 1 0.15 008
PGI2 l000nM 45.2 7.5 9.1 3.5 0.9 0.4 32.2 8.Ot 237 4.3 0.05 002

Studies performed on citrated blood from each of four normal subjects C. S. and T denote percentage surface coverage
with contact platelets (Cl. spread platelets(S). and thrombi (T(. lOOC/(C+ 5) reflects the ability of contact platelets to spread
on the subendothelium. Statistical significance of C + S values (compared with vehicle( are shown NS. not significant:
p < O.05.S.p <0.01 (Studentsttest(
Values are the mean SE

Table 2. Platelet Interaction With a-Chymotrypsin.-Digested Subendothelium

(Fibrillar Collagen)
Shear Rate 800 sec 10 mm

C+S lOOT/S
Addition (Adhesion( (%( 1 OOC/C + S (Aggregation(

Nothing 22.9 2.6 3.2 0.7 59.1 8.7

Vehicle 24.1 4.8 1.3 0.3 56.6 7.5
PGI2 lOnM 16.1 3.2(NS) 4.7 1.4 6.1 2.8
PGI2 100nM 9.7 3.8(NS) 5.8 2.9 2.5 1.1
PGI21000nM 4.1 3.3 7.1 4.2 0

Studies and symbols as in Table 1.

eliminated platelet thrombi. However, this surface is considerably less thrombo-

genie (control lOOT/S values only 9%-I5%) than the fibrillar collagen surface of
a-chymotrypsin-digested subendothelium.
PGI, also inhibited platelet adhesion to subendothelium. This was best demon-
strated by the studies performed at the higher shear rate (2600 sec). As seen in
Table I, addition of PGI, to blood in concentrations of 10, 100, and 1000 nM
resulted in a significant reduction of platelet adhesion values from 75.9% 4.1%
(observed with vehicle-treated blood) to 40.2% 7.8%, 33.6% 4.0%, and 3 2.2%
8.0%, respectively. At the lower shear rate (800 sec), adhesion was reduced
from 79.7% 8.6% to 65.9% 4.2%, 54.8% 13.1%, and 45.2% 7.5% by
these same concentrations of PGI2 (only the last reduction was significant). The
adhesion of platelets to the fibrillar collagen surface of a-chymotrypsin-digested
subendothelium was also inhibited by PG!2 (Table 2).

DISCUSSION
We studied the interaction of platelets in flowing blood with subendothelium at
shear rates comparable to those in the arterial circulationt and found that PGI2
markedly inhibited thrombus formation (lOOT/S). This result was achieved at a
concentration of PGI2 (10 nM) that also inhibited platelet-to-platelet interaction
(aggregation) induced in the aggregometer by ADP and a concentration of collagen
that induces platelet aggregation that is inhibited by aspirin. It is currently held
that another prostaglandin, PGEI, inhibits platelet aggregation and shape change
by stimulating adenyl cyclase, thereby increasing the basal levels of platelet
From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

PROSTACYCLIN INHIBITS PLATELET ADHESION 249

cAMP.92#{176} Since PGI2 is an even more potent stimulator ofcAMP accumulation in

platelets,2122 its inhibitory effects on platelet-to-platelet interaction (e.g., aggrega-
tion and thrombus formation) could occur through a similar mechanism.
The results of the study also demonstrate that P012 in a concentration of 10 nM
inhibits platelet adhesion (C + 5) to subendothelium. However, at no concentra-
tion was adhesion completely eliminated, and, as also reported by Higgs et al.,24 the
inhibitory effect of PGI2 on platelet adhesion was less pronounced than its effect on
thrombus formation. The mechanism by which PG!2 inhibits platelet adhesion
requires further study. As previously observed with PGEI,23 the values of lOOC/
(C + S) are increased by PGI2 (Tables I and 2). This expression reflects the ability
of platelets that have made contact with the subendothelium to change their shape
and spread upon this surface. The increased values of 100C/(C + S) observed with
PGI2 indicate that it inhibits platelet spreading, perhaps by the same mechanism
that accounts for its inhibitory effect on the shape change induced by aggregating
agents (Fig. 1). Since platelets that have initially attached to a surface, but have
not spread upon it, are more likely to be removed by the high shear rates at the
vessel wall, it is possible that PGI2 facilitates the removal of contact platelets by
preventing their spreading. This might also explain the greater inhibitory effect of
PGI2 at the higher of the two shear rates used in this study and the failure of Higgs
et al. (who used only the lower shear rate) to observe inhibition of adhesion except
at PGI, concentrations of 50 nM and greater.24 In the present study, we found that
10 nM PGI2 inhibited platelet adhesion by about 45% when studied at a shear rate
of 2600 sec, and it might be an even more potent inhibitor of adhesion at higher
shear rates. This could be of significance in view of recent studies indicating that
shear rates as high as I 6,000 sec may exist in the microvasculature.2526 The shear
rate dependence of the platelet adhesion defect produced by PGI2 is reminiscent of
similar observations in patients with von Willebrand disease. In addition, this
finding suggests that PGI,, either alone or through its synergistic effects on drugs
such as dipyridamole,27 may be a more effective inhibitor of platelet adhesion in the
arterial (high shear) than in the venous (low shear) circulation. Finally, since we
used deendothelialized blood vessels, the results of the study are not directly
applicable to the question of platelet-endothelial interactions (or lack thereof),
although inhibition of platelet spreading on surfaces could well be a general
property of PG 12.
It is currently held that removal of the endothelium by a variety of mechanisms
(immunologic, toxic, hemodynamic) is followed by the deposition of platelet
thrombi and release from the platelets of biologically active substances. The results
of this study suggest that PG!2 (produced either in adjacent endothelial cells or in
other cells of the vessel wall) could play a role in regulating the deposition of
platelets on the subendothelium. Hence PG!2 could be an important regulator of a
variety of platelet-mediated biologic processes.

ACKNOWLEDGMENT
We thank John Rogers, Sachi Senrui, and Thomas Hoffmann for skillful technical assistance.

REFERENCES
I. Baumgartner HR: The role of blood flow in of mural thrombi. Microvasc Res 5:167-179,
platelet adhesion, fibrin deposition and formation 1973
From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

250 WEISS AND TURITTO

2. Samuelsson B: The role of prostaglandin ing disorders. Thromb Haemostasis 37:17-28,

endoperoxides and thromboxanes in human I977
platelets. in Silver MI, Smith JB, Kocsis JJ (eds): I 5. Tschopp TB, Weiss Hi, Baumgartner HR:
Prostaglandins in Hematology. New York, Spec- Interaction of platelets with subendothelium in
trum. 1977, pp 1-10 thrombasthenia: Normal adhesion, impaired ag-
3. Ross RI, Glomset J, Kariya B, Harker LA: gregation. Experientia 31:1 13-I 6, 1975
A platelet-dependent serum factor that stimulates 16. Weiss Hi, Tschopp TB, Baumgartner HR.
the proliferation ofarterial smooth muscle cells in Sussman II, Johnson MM, Egan II: Decreased
vitro. Proc NatI Acad Sci USA 71:1207-1210, adhesion of giant (Bernard-Soulier) platelets to
I 974 subendothelium-Further implications on the role
4. Holmsen H: Biochemistry of the platelet of the von Willebrand factor in hemostasis. Am I
release reaction. Ciba Found Symp 35:175-196, Med 57:920-925, 1974
I 975 17. Nicolaou KC, Barnette WE, Gasic GP,
5. Weiss HI: Antiplatelet therapy. N EngI I Niagolda RL, Sipio Wi: Simple efficient synthesis
Med 298:1344-1347, 1403-1405, 1978 of prostacyclin (PGI2). I Chem Soc Chem
6. Moncada 5, Higgs EA, Vane IR: Human Commun 18:630-631, 1977
arterial and venous tissues generate prostacyclin 18. Weiss Hi, Aledort LM, Kochwa S: The
(prostaglandin X), a potent inhibitor of platelet effect of salicylates on the hemostatic properties
aggregation. Lancet l:l82l, 1977 of platelets in man. I Clin Invest 47:2169-2179,
7. Moncada S, Herman AG, Higgs EA, Vane 1968
IR: Differential formation of prostacyclin (PGX 19. Shio H, Ramwell PW, lessup SI: Prosta-
or PCI2) by layers of the arterial wall. An expla- glandin E: Effects on aggregation, shape change
nation for the anti-thrombotic properties of vascu- and cyclic AMP of rat platelets. Prostaglandins
ar endothelium. Thromb Res I I :323-344, 1977 1:29-35, 1972
8. Iohnson RA, Morton DR. Kinner IH, 20. Salzman EW: Cyclic AMP and platelet
Gorman RR, McGuire JC, Sun FF, Wittaker N. function. N EngI I Med 286:358--363, 1972
Bunting S. Salmon I, Mbncada 5, Vane IR: The 21. Tateson iE, Moncada S. Vane IR: Effects
chemical structure of prostaglandin X (prostacy- of prostacyclin (PGX) on cyclic AMP concentra-
din ). Prostaglandins I 2:9 1 5-928, 1976 tions in human platelets. Prostaglandins 13:389-
9. Weksler BB, Marcus AJ, Jaffe EA: Synthe- 398, 1977
sis of prostaglandin 12 (prostacyclin) by cultured 22. Gorman RR. Bunting S. Miller OV:
human and bovine endothclial cells. Proc Natl Modulation of human platelet adenylate cyclase
Acad Sci USA 74:3922-3926, 1977 by prostacyclin (PGX). Prostaglandins 13:377-
10. Baenziger NL, Dillender MI, Majerus 388, 1977
PW: Cultured human skin fibroblasts and arterial 23. Baumgartner hR. Muggli R, Tschopp TB.
cells produce a labile platelet-inhibitory prosta- Turitto VT: Platelet adhesion, release and aggre-
glandin. Biochem Biophys Res Commun 78:294- gation in flowing blood: Effects of surface proper-
301. 1977 ties and platelet function. Thromb Haemostasis
11. Weiss HI, Turitto VT, Baumgartner HR:
35:124-138, 1976
Effect of shear rate on platelet interaction with
24. Higgs EA, Moncada S. Vane IR, Caen IP,
subendothelium in citrated and native blood. I.
Michel H, Tobelem G: Effect of prostacyclin
Shear-dependent decrease of adhesion in von
(PGI2) on platelet adhesion to rabbit arterial
Willebrands disease and the Bernard-Soulier
subendothelium. Prostaglandins 16:17-22, 1978
syndrome. I Lab Clin Med 92:750-764, 1978
25. Copley AL, King RG: Polymolecular
12. Tschopp TB, Baumgartner HR. Silber-
layers of fibrinogen systems and the genesis of
bauer K, Sinzinger H: Platelet adhesion and
thrombosis. Thromb Res 8 [Suppl 2]:393-408.
platelet thrombus formation on subendothelium of
1976
human arteries and veins exposed to flowing blood
in vitro. Haemostasis (in press) 26. Schmid-Schonbein GW, Zweifach BW:
13. Weiss Hi, Baumgartner HR. Tschopp TB, RBC velocity profiles in arterioles and venules in
Turitto VT: A new method for identifying and the rabbit omentium. Microvasc Res 10:153-164,
classifying abnormalities of platelet function. Ann 1975
NY Acad Sci 283:293-309, 1977 27. Moncada 5, Korbut R: Dipyridamole and
14. Baumgartner HR. Tschopp TB, Weiss Hi: other phosphodiesterase inhibitors act as anti-
Platelet interaction with collagen fibrils in flowing thrombotic agents by potentiating endogenous
blood. II. Impaired adhesion-aggegation in bleed- prostacyclin. Lancet 1:1286-1289, 1978
 From www.bloodjournal.org by guest on November 7, 2017. For personal use only.

1979 53: 244-250

Prostacyclin (prostaglandin I2, PGI2) inhibits platelet adhesion and

thrombus formation on subendothelium
HJ Weiss and VT Turitto

Updated information and services can be found at:

http://www.bloodjournal.org/content/53/2/244.full.html
Articles on similar topics can be found in the following Blood collections

Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

Information about ordering reprints may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of
Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.