British Journal ofhlaernutoiogy, 1979, 43, 637-647.

Prostaglandins as Inhibitors of Human Platelet Aggregation G. DI MINNO,M. J. SILVER AND G. DE GAETANO Irtituto di Ricerche Farmacologiche ‘ M a r i o Negri’, M i l a n o , Italy (Received 4 December 1978; accepted for publication 22 February 1979) ARY. The potencies of prostaglandins (PG) I,, PGD, and PGEl as inhibitors of S human platelet aggregation induced by threshold concentrations of four aggregating agents were determined in platelet-rich plasma from normal individuals who had not ingested aspirin. The order of activity against ADP, adrenaline and collagen was always PGI, > PGD2 > PGE1. However, PGD, and PGEl were almost equipotent with PGI, when tested against arachidonic acid (AA). The threshold inhibitory effects of PGD,, PGEl and PG12 could be overcome by increasing the concentrations of the aggregating agents AA, collagen or ADP. Adrenaline was found to be different from the other aggregating agents. It could overcome inhibition of platelet aggregation by PGD, but could not overcome inhibition by PG12 or PGE1. These facts support the hypothesis that platelet receptors for PG12 and PGEl are similar to each other and different from the receptor(s) for PGD,. PRP obtained from normal subjects after the ingestion of aspirin exhibited only one wave of aggregation in response to ADP, adrenaline or collagen. PG12, PGD, and PGEl were all powerful inhibitors of this single wave of aggregation. The inhibitory activity of all three prostaglandins a t threshold concentrations was overcome by increasing the concentration of ADP or collagen but not by increasing the concentration of adrenaline.

Three naturally occurring prostaglandins are potent inhibitors of platelet aggregation: prostaglandin (PG) El (Kloeze, 1967), PGD, (Smith et al, 1974) and PGI, (Moncada et al, 1976). While their precise mechanisms of action are unknown, all increase intracellular levels of C-AMP (Robison et all 1969; Mills & Macfarlane, 1974; Best et al, 1977; Gorman et al, 1977; Tateson et al, 1977). The receptor for PGD, on human platelets is apparently different from that for PGI, and PGEl (Mills et al, 1977; Siegl et al, 1978; Whittle et all 1978; MacIntyre & Gordon, 1977). Because slowly metabolized analogues of the prostaglandins might have antithrombotic properties and because it is essential to clarify their likely physiological role in haemostasis, we studied their effects on the aggregation of normal human platelets. This is the first detailed study in which all three prostaglandins are directly compared using four different aggregating agents at threshold aggregating concentrations. We also describe studies on Correspondence: Dr M. J. Silver, Cardeza Foundation, Thomas Jefferson University, 1015 Walnut Street, Philadelphia, Pa. 19107, U.S.A. 0007-1048/79/1200-0637 $02.00 01979 Blackwell Scientific Publications

637

638

C . Di Minno, M . J . Silver and G. de Caetano

inhibition of platelet aggregation by the three prostaglandins in platelet-rich plasma (PRP) obtained from individuals who had ingested aspirin. MATERIALS A N D METHODS

Agqre'qating agents employed were AIIP (adenosine-5'-diphosphate, sodium salt), Sigma Co., St Louis, Missouri; adrenaline bitartrate, Mascia-Brunelli, Milan, Italy; collagen (Kollagen reagents Horm), Hormon Chemie, Munchen; arachidonic acid (AA) (>99"/0),Nuchek, Elysian, Minnesota. Solutions of the sodium salt were prepared as described elsewhere (Silver et al, 1973). Dilutions of the agents for aggregation tests were in tris buffer (0.15 M , pH 7-4). Prosta'qlandins. Synthetic PG12, PGEI and PGD, were gifts from Dr John Pike of the Upjohn Co., Kalamazoo, Michigan. The prostaglandins were dissolved in ethyl alcohol and these stock solutions (1-3 mM) were stored a t -20°C and diluted in tris buffer (pH 9.0 for PG12 and pH 7.4 for PGE, and I'GD,) just before each series of tests. Aspirin (Bayer, Milan, Italy). A 500 mg tablet was ingested by each healthy volunteer 1.5-2 h before taking blood samples. Platelet-rich plasma. Nine volumes of blood, collected from the antecubital veins of healthy volunteers, were mixed with one volume of 3.8% trisodium citrate. The volunteers denied taking any drugs for a t least 10 d before donating blood. The blood was immediately centrifuged a t 200 g for 15 min a t room temperature (20-22°C) and the PRP was removed. Platelet counts on PRP were between 300 and 400 x 109/l. Neither blood nor PRP contacted glass surfaces. Plarelet-aRRre~ati(1Mtests. All tests were performed between 1 and 3 h after venipuncture. Aggregation tests were done in plastic cuvettes in the Elvi-840 aggregometer, Elvi-Logos, Milan, Italy. Inhibitors, or an equal volume of vehicle, were added in microlitre amounts to 0.25 ml of PRP which had been stirred at 1000 rpm and at 37°C for 1 min before adding the aggregating agent. DEFl N I TIONS

A. Studies in which Aspirin was not Employed For each aggregating agent and for each PRP studied the Threshold ABregating Concentration (TAC) was defined as the smallest amount of aggregating agent producing more than 80% light transmittance for collagen and arachidonic acid and causing two waves of aggregation for ADP and adrenaline. Threshold Inhibitory Concentration (TIC) was defined as the smallest amount of each PG causing more than 90% inhibition of aggregation induced by each aggregating agent at the threshold aggregating concentration. The overcoming concentration (OC) was defined as the smallest amount of aggregating agent necessary to overcome the inhibitory effect obtained a t the threshold inhibitory concentration. The inhibitory effect was considered to be overcome when the aggregation recording obtained was similar to that seen at the threshold aggregating concentration in the absence of inhibitor. B. Studies in which Aspirin was Employed Aspirin inhibits the platelet release reaction: after aspirin ingestion only one wave of

Platelet A ~ r e ~ ~ a t i o n - P r o s t a ~ ~ l a nInhibitors din

639

aggregation occurs in PRP in response to aggregating agents like adrenaline and ADP. Because of the smaller aggregation responses in PRD obtained from subjects who had taken aspirin the above definitions were modified for the aspirin studies. The threshold aggregating concentration for collagen was defined as the lowest concentration causing more than 40%0 light transmittance within 2 min and that for A D P and adrenaline a s the smallest concentration causing one wave of aggregation with a t least 25% light transmittance. PRP from each individual who had ingested aspirin was tested for aggregation in response to AA. Platelet aggregation never occurred a t concentrations as high as 1 mM and therefore AA was not used further in the aspirin studies. In other studies, platelets in PRP obtained after ingestion of aspirin took up [14C]hydroxytryptaniinenormally but did not release it in response to the aggregating agents employed here (PRP from 12 different subjects, data not shown). Inhibitory effects seen in studies on PRP obtained after ingestion of aspirin were considered to be similar to inhibition of the first wave of aggregation normally seen in response to low concentrations of AIIP or adrenaline in thc absence of aspirin. This aggregation is also referred to as aggregation without release. In all cases when TAC, TIC or OC were determined for a particular PRP, appropriate controls were performed just before and after the critical test. Also, controls were run rcpeatedly during a series of tests on the same day and on PRP obtained from the same individual on different days to ascertain that solutions of the PGs were not losing inhibitory activity. RESULTS

Studies on PRPfvorn Normal Individuals who had not Ingested Aspirin Relative Potencies ‘$&‘GI2,PGD, and PGEl as Inhibitors of Aggregation of Human Platelets In all cases (Fig 1) the mean TIC for PG12 was less than 20 nM. PGD2 and PGEl were almost equipotent with PGI, as inhibitors of aggregation induced by AA. PGD2 was 3 times less potent than PGI, in inhibiting aggregation induced by collagen and about 5-6 times less potent in inhibiting aggregation induced by ADP or adrenaline. PGE, also strongly inhibited aggregation induced by collagen (TIC, 25 nM) but was about 10-14 times less potent than PG12 in inhibiting aggregation induced by ADP or adrenaline. T he mean potencies of PG12 relative to PGD, or PGEl calculated from the data presented in Fig 1 were respectively as follows: against AA, 1.3, 1.6; against collagen, 3.2,6.2; against ADP, 5,5, 13.5; against adrenaline, 6.4, 10.5. Adrenaline Overcomes Inhibition of Platelet Aggregation by PGD2 but not by PG12 or P G E , After finding the threshold aggregating concentration of the four aggregating agents on each PRP, the T I C for each prostaglandin was determined. Then, the OC for each aggregating agent against each prostaglandin was found (for example see Fig 2B, where adrenaline a t 100 PM overcomes the inhibitory effect of PGD,). The results are given in Table I. It can be seen that large excesses of ADP were needed to overcome the inhibitory effects of the prostaglandins whilst intermediate quantities o f collagen were suficient and relatively small amounts of AA sufficed.

C . Di Minno, M . J . Silver and C . de Caetano

ARACHIOONIC ACID

COLLAGEN

ADP

ADRENAL INE

FIG I . Inhibition ofplatelet aggregation by PG12, PGD2 and PGEl against four different aggregating agents. The heights of the columns are the mean values for threshold inhibiting concentration (TIC). Dots represent individuals values and indicate the number of determinations performed with PRP from different normal individuals who had not ingested aspirin. SEM are indicated.

4

SALINE

PG D2

W

v

z a I-

c -

I v)

z a a

I-

c

I

!?-I

s

PGEl PGl2

FIG 2. (A) Recordings of aggregation induced by adrenaline a t threshold aggregating concentration and inhibition produced by PGEI, PGDl and PGIz. (B) Recordings of aggregation observed when adrenaline was added in high concentration to overcome the inhibitory effects of the prostaglandins studied.

Inhibitors

Platelet Agqregation-Prostaglandin

641

TABLE I. Range of values obtained for threshold aggregating concentration (TAC) and overcoming concentration (OC) for aggregating agents. Studies on PRP obtained from normal subjects who had not ingested aspirin.

O C (range) Aggregating agent

TAC (range)

ADP (PI)

0.25-4.0 n*=16 Collagen (pglml) 0. 64. 0 n=21 Arachidonic acid (mM) 04-0.6 n=8 Adrenaline (PM) 2.5-20 n=25

PCI,

PCDZ

PCEl

8-40 n=5 2-20 n=7 0.8-1 n=7

4-40 n=5

4-40 n=6

4-16 n=5 0.8-1 n=7

4-16 n=5 0.8-1 n=8

Does not over comet n=5

40-200

Does not overcome n=10

n=8

* n=Number of determinations. PRP from 23 different individuals was used in these studies. t Adrenaline was always tested at concentrations up to 500 PM. When adrenaline was the aggregating agent, the inhibitory effect of PGD2 could be overcome by 10-16 times the TAC (Table 11). However, adrenaline could not overcome the inhibitory effects of PG12 or PGE1, even at concentration as high as 500 p~ (Fig 2). Tracing A shows inhibition of adrenaline (5 p ~induced ) aggregation by PGE1, PGD2 and PG12. While curve B shows that adrenaline can overcome the inhibition caused by 125 nM PGD2 but not that caused by 30 nM PG12 or 350 nM PGEI. Similar results were obtained in seven experiments. This experiment is shown because of the high concentrations of inhibiting prostaglandins employed. TABLE 11. Ratio of OC to T A C for ADP, arachidonic acid, collagen and adrenaline. Studies on PRP from normal subjects who had not ingested aspirin.

oc ~

TAC

(range) withfollowing aggregating agent:

Inhibitor

ADP

Collagen Arachidonic acid Adrenaline

PGIz PGDz PGEl

10-16* 4.04.6 10-32 3'3-5.0 10-16 4'04.6

1'62.0 1.62.0 1.6-2'0

-t 10-16

-t

* Values shown here were calculated from the data for respective OCs and TACs of Table I. t Ratio could not be determined because there was no oc.

642

G. Di Minrzo, M . J . Silver and G. de Gaetarzo

Studies on P R P Obtainedfvom Normal Subjects after the InAestion o f Aspirin Typical aggregation recordings a t the T A C for ADP, adrenaline and collagen on 1’RP obtained after the ingestion of aspirin are shown in Fig 3 which also shows a typical recording in response to 1 mM AA (no aggregation). PRPs from 12 differcnt individuals who had ingested aspirin were testcd in similar fashion. The aggregation tracings in response to 1 mM AA were similar to that seen in Fig 3.

W

Collagen I l p g / m l )

50

u

5 LO

ADP ( 2 pM 1

I-

t

3;

30

ADREN ( 2 0 pM J

z

Q:

20 I-

I

12 10

AA

-I

8

SALINE

0

G, 0

(1mM)

AGGREGATING AGENT I

1

2 3 MINUTES

1

5

FIG 3. Recording5 of platelet aggregation obtained in response to threshold aggregating concentrations of ADP, collagen and adrenaline in PRP obtained from healthy individuals, 1.5-2 h after ingestion of 500 mg aspirin. Also shown is a recording obtaincd after additions of 1 m M (final concentration) AA to PRP from an individual who had ingested aspirin.

Inliibitiun of Aggregation without Release b y P C I z , PGD2 and PCEl. All three prostaglandins are powcrful inhibitors of aggregation without release. The mean T A C for each was between 14 and 97 nM against the three aggregating agents used (Fig 4). Against all three aggregating agents tested PGIZ was between 2.1 and 3.3 times as potcnt as PGD2 and betwecn 2.7 and 5.3 times as potent as PGEl (see Table 111 for summary of relative potencics).

ADP and Collagen can Overcome the Inhibitory Eflects ofPGI2,PGD, and PGEl in P R P Obtained after Ingestion of Aspirin but Adrenaline can not To explore the mechanisms involved in the inhibitory activity of these three agents on the first wave of platelet aggregation, we tested the ability of aggregating agents to overcome the inhibition produced by the three PGs. Table IV shows ranges of T A C and OC obtained for ADP, collagen and adrenaline. Table V gives the ratio of O C / T A C for the three aggregating agents. Collagen overcame the inhibitory effect of the three PGs at concentrations 2.5-10 times the TAC, whilst ADP required 5-30 times as much. O n the other hand, adrenaline could not overcome the inhibitory effects of the PGs a t concentrations as great as 500 times the T A C (Table IV).

Platelet Aaregation-Prostaglandin

Inhibitors

643

TABLE 111. Mean relative potencies of PGIz, PGD2 and PGEl as inhibitors of first wave of platelet aggregation.* Studies on PRP from normal subjects after ingestion of aspirin. Putency ufPG12 relative to: Aggregating agent PCD2 PCEl ADP Collagen Adrenaline

2.1 3.3 2.8

2.7 5.3 2.5

* Values shown here were calculated from the mean TICSofFig 4.

..

ADP

COLLAGEN

ADRENALINE

FIG4. Inhibition of platelet aggregation by PG12, PGDz and PGEi against three differeht aggregating agents. The heights of the columns are the mean values for threshold inhibitory concentrations (TIC). Standard error of the mean is shown in each case. Dots represent individual values and indicate the number of determinations performed with PRP (obtained after ingestion of aspirin) from different individuals.

DISCUSSION Studies on Normal Human Platelet-Rich Plasma Relative potencies of inhibitors of aggregation depend on the aggregating agent employed. Assessing the relative potencies of these prostaglandins as inhibitors of aggregation must take into account the particular aggregating agent used (Fig 1).For example, PGIz is more than 6 times

G. Di Minno, M . J . Silver and G. de Gaetano

644

TABL IV.~ Range of values obtained for T A C and OC. Studies on PRP obtained from normal subjects after ingestion of aspirin.

OC (range) Amregating agenf

TAC (ra n'v)

PGIZ

PGD2

PCEl

15-40 8-20 n=4 n=4 n=4 Collagen (pg/ml) 0.5-2 3-5 3-5 3-20 n=10 n=4 n=4 n=4 Adrenaline ( p ~ ) 1-40 Does not overcomet Does not overcome Docs not overcome n=9 n=9 n=h n=h

ADP (PM)

0.5-4 n*=6

10-40

* n=number of determinations. PRPs from 10 different individuals were used in these studies. t Adrenaline was always tested at concentrations up to 500 PM.

TABLE V. Ratio of OC to T A C for ADP, collagen and adrenaline. Studies on PRP from normal subjects after ingestion of aspirin. OCI TAC (range) with following aggregating agent: Inhibitor Collagen ADZ' PGIZ PGDZ PGE1

2.54* 2.5-6 610

1C20 10-30 5-16

Adrenaline

-t -t -t

* Values shown here were calculated from the data for respective O C s and TACs of Table IV. t Ratio could not be determined because there was n o O C . as potent as PGD2 in inhibiting aggregation induced by adrenaline but only 1.3 times as strong as PGD2 when tested against AA. Should one conclude that PGI, is equipotent with o r 6 times as strong as PGD, as an inhibitor of platelet aggregation? Clearly, the relative potencies of

inhibitors of platelet aggregation should always be qualified by reference to the particular aggregating agent used. The data showing that PGI, is about 6 times as potent as PGD, and about 14 times as potent as PGEl in inhibiting aggregation induced by ADP (Fig 1)differ from the findings of Whittle et a2 (1978) who report that PG12 is 20 times as potent as PGD, and 50 times as potent as PGE1. Our findings show that the average inhibitory concentration for all the prostaglandins

Platelet Aggregation-Prostaglandin

Inhibitors

645

varied between 4 and 160 nanomolar (Fig 1). Thus, all thrce prostaglandins appear to be excellent inhibitors of platelet aggregation. Platelet receptors for aggregating agents and prostaglandins that inhibit aggregation. PGD2 is a potent inhibitor of the aggregation of human platelets but not of rat or rabbit platelets (Smith et al, 1974) and PGEl inhibits the aggregation of rat platelets while PGD2 did not (Kloeze, 1967). Furthermore PGI2, like PGEl inhibited aggregation of sheep, horse, human, rabbit and rat platelets (Whittle et al, 1978). These differences between and similarities of the effects of inhibitors on platelets from different species suggest that there may be corresponding similarities and differences in receptors on the platelets from the various species. The similarities and differences between PGE1, PG12 and PGD, reported here are in accord with recent preliminary findings concerning their interaction with platelet receptors that regulate adenylate cyclase activity and studies on the binding of radioactive PG12 to platelet receptors. Mills et a1 (1977) reported that PGEl inhibited the formation ofC-AMP by human platelets in response to PG12 while PGD2 did not. Siegl et a1 (1978) showed that PGE, could inhibit the binding of radioactive PG12 to its high affinity binding site, but PGD2 was unable to do so, even at very high concentrations. MacIntyre & Gordon (1977) showed that the compound N-0164 antagonized the inhibitory effect of PGD,, but not that of PGE1. Thus, our findings on the ability of adrenaline to overcome the inhibitory activity of PGD2 but not that of PGEl or PG12 support and extend previous reports indicating that receptors for PGD2 are different from those for PGEl or PG12. Znhibition of arachidonic acid-induced aggregation. One of the most interesting findings in this study is that PGD2 and PG12 are equipotent as inhibitors of AA-induced aggregation and not much more potent than PGE1. Thus, at concentrations of about 10 nM the three PGs completely inhibited aggregation induced by AA (Fig 1). Furthermore, the inhibitory effects of all three PGs may be overcome by relatively small additions of exogenous AA (see Table 11). Since AA-induced aggregation is essentially related to the release reactioq, these findings suggest that the three PGs inhibit the second wave of aggregation by a similar mechanism. Studies on Human Platelet Rich Plasma Obtained after the Ingestion o f Aspirin Suppression ofsecond wave ofaggregation to studyfirst wave. We were able to examine the effects of these inhbitors on the first wave of aggregation alone by suppressing the formation of the second wave. The validity of the approach used here to study effects of inhibitors on the isolated first wave of aggregation may be questioned. For example, whilst a major effect of aspirin on human platelets was shown to be acetylation of the enzyme cyclooxygenase (Roth & Majerus, 1975) other proteins were also shown to be acetylated and it is not known whether ingested aspirin may affect other platelet constituents. T o study the first wave we considered using concentrations of ADP or adrenaline which would give only one wave aggregation when added to normal human PRP without aspirin. Although this approach has been used by Crane et al (1978) we did not use it because OC of aggregating agents always produced a second wave. Suppression of the second wave of aggregation by aspirin provides conditions for studying the first wave of human platelet aggregation in isolation. Potency ofPGI2,PGD2 and PGEl in inhibiting thejrst wave ofaggregation. The three PGs were found to be potent inhibitors of the first wave of aggregation, all causing complete inhibition of aggregation in nanomolar concentrations (Fig 4).

646

G. D i Minno, M . J . Silver and G. de Caetano

Inhibition of isolatedfirst wave o f aggregation compared to inhibition o f combined first and second waves. It might be expected that more inhibitor would be necessary to inhibit the combined first and second waves of platelet aggregation than would be necessary for the isolated first wave, but we found the opposite to be true. While the TIC against first wave aggregation for any of the prostaglandins was never below 10 nM (Fig 4), for inhibition of combined first and second waves 50% of the TICs were 10 nM or less (see Fig 1).This seeming paradox may be explained by the fact that cyclooxygenase inhibitors can also suppress the formation of endogenous PGDz by platelets (Oelz et al, 1977).Thus without aspirin this endogenous PGD2, formed during aggregation, may act in concert with the exogenous prostaglandin inhibitors and so less exogenous inhibitor would be necessary. The possibility that endogenous PG12 in plasma might explain this paradox is unlikely because we always allowed PRP to stand for 1 h after venepuncture before use in aggregating tests. This plasma would not contain the unstable PGI2 which would be rapidly converted to the inactive metabolite 6-keto-PGFt, (Johnson et a l , 1976). Inhibition of aggregation without release (first wave) compared to inhibition .f aBregation with release (second wave). Whilst the threshold inhibitory concentrations of the three PGs against the first wave of aggregation (Fig 4) were always above 10 nM, 60% of the TICs against the second wave were below 10 nM (TICs for AA, Fig 1). Thus, the second wave of aggregation appears to be more easily inhibited by these prostaglandins than the first wave and suggests quantitative or qualitative differences between the platelet receptors controlling each of these waves of aggregation. It is also possible that aspirin may modify the receptors for first wave aggregation slightly, so that more inhibitor is required to suppress it than in the absence of aspirin. Adrenaline can not overcome inhibition ofthejrst wave of aggregation b y PGIZ, PCDz or PGE,. O u r finding that adrenaline could overcome the inhibitory activity of PGDz when tested on normal PRP but could not overcome inhibition caused by PG12 and PGE, suggested that platelet receptors of PG12 and PGEl were similar to each other and different from those of PGDz. However, in PRP obtained after the ingestion of aspirin, adrenaline could not overcome the inhibiting activities of any of the prostaglandins against the isolated first wave of platelet aggregation. This suggests that there are differences in platelet receptors for adrenaline influencing first wave aggregation and those influencing second wave aggregation. Thus the prostaglandin receptors for first wave aggregation are similar in that all three PGs inhibit completely and irreversibly the first wave of aggregation induced by adrenaline. Prostaglandin Iz has been called the most powerful inhibitor of platelet aggregation yet discovered (Moncada et al, 1976) but we would emphasize that it is one of three, potent, naturally occurring prostaglandins whose physiological significance needs to be evaluated. Since human platelets can produce significant amounts of PGDz (Oelz et al, 1977) and blood vessels can produce PG12,any hypothesis referring to significancein haemostasis or thrombosis must take into account these two potent inhibitors of platelet aggregation. PGE1, which is not formed by platelets or blood vessels, is also a potent inhibitor of aggregation and its possible role in platelet physiology and pathology can not, as yet, be dismissed. Analogues of all three of these prostaglandins may prove to be valuable antithrombotic agents.

Platelet Aggregation-Prostaglandin

Inhibitors

647

REFERENCES L.C., MARTIN, T.J.,RUSSELL, R.G.G. & PRESF.E. (1977) Prostacyclin increases cyclic AMP lcvels and adenylate cyclase activity in platelets. Nature, 267, 85(t851. CRANE, H.,MAIsH,T.L.,MADDox, Y.T.,COREY,E.J., SZEKELY, T. & RAMWELL, P.W. (1978) Effect of prostaglandin I2 and analogs on platelet aggrcgation and smooth muscle contraction. Journal of‘ Pharmacology and Experimental Therapeutics, 206, 132-1 38. GORMAN, R.R., BUNTING, S. & MILLER, O.V. (1977) Modulation of human platelet adenylate cyclase by prostacyclin (PGX). Prostagfundins, 13, 377-388. JOHNSON, R.A., MORTON, D.R., K1NNER.J.H.. GORMAN, R.R., MCGuIRE,J.C., SUN, F.F., WHFITAKER, N., BUNTING, S., SAI.MON, J., MONCAIIA, S. & VANE,J.R. (1976) The chemical structure of prostaglandin X (prostacyclin). Prostaglandins, 12, 915-928. KLOEZE, J. (1967) Influence of prostaglandins on platelet adhesiveness and platelet aggregation. Prostaglandins (ed. by S. Gerstrom and B. Samuelsson), pp. 241-252. Intexcience, London. J.L. (1977) DiscriminaMACINTYRE, D.E. & GORDON, tion between platelet prostaglandin receptors with a specific antagonist of bisenoic prostaglandins. Thrombosis Research, 11, 705-713. MILLS, D.C.B. 81 MACFARLANE, D.E. (1974) Stimulation of human platelet adenylate cyclase by prostaglandin D,. Thrombosis Research, 5 , 40-412. MILLS,D.C.B., MACFARLANE, D.E. & NICOLAOU, K.C. (1977) Interaction ofprostacyclin (PG12)with thc prostaglandin receptors on human platelets that regulate adenylate cyclase activity. Blood, 50, Suppl. 1, 247. S., GRYGLEWSKI, R., BUNTING, S. & VANE, MONCADA, BEST,

TON,

J.R. (1976) An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature, 263, 663-665. OELZ, O., OELZ, R., KNAPP, H.R., SWEETMAN, B.J. & OATES, J.A. (1977) Biosynthesis of prostaglandin DZ. Prostaglandins, 13, 225. R.C. ROBISON, G.A., ARNOLD,A. & HARTMANN, (1969) Divergent effects of epinephrine and prostaglandin El on the level of cyclic AMP in human blood platelets. Pharmacological Research Communications, 1, 325-329. hni, G.J. & MAJERUS, P.N. (1975) The mechanism of thc cffect of aspirin on human platelets. I. Acetylation of a particulate fraction protein. journal of Clinical Znvesttqation, 56, 624-632. SIEGL, A.M., SMITH,J.B., SILVER, M.J., NrcoLAou, W.E. (1978) Binding K.C., GASIC, G. & RURNETTE, of prostacyclin by platelets. Federation Proceedings, 37,260. SILVER, M.J., SMITH, J.B., INGERMAN, C . & KOCSIS, J.J. (1973) Arachidonic acid-induced human platelet aggregation and prostaglandin formation. Prostuglandins, 4, 863-875. SMITH, J.B., SILVER, M.J., INGERMAN,C.M. & KOCSIS, J.J. (1974) Prostaglandin D2 inhibits the aggregation of human platelets. Thrombosis Research, 5, 29 1-29 9. TATESON, J.E., MONCADA, S. & VANE,J.R. (1977) Effects of prostacyclin (PGX) on cyclic AMP concentrations in human platelets. Prostaglandins, 13, 389-397. WH~LE B.J.R., , MONCADA, S. & VANE,J.R. (1978) Comparison of the effects of prostacyclin (PG12), prostaglandin E l and D2 in platelet aggregation in different species. Prostaglandins, 16, 387-402.