1189

Platelet-Activating Factor (PAF) Receptor Antagonists Inhibit Arachidonic Acid Induced Platelet Aggregation in Rabbit Whole Blood' Alaina Jean Ammit and Chris O'Neill* Human Reproduction Unit, Royal North Shore Hospital of Sydney, St. Leonarcls, 2065, N.S.W., Australia

The potency of several platelet-activating factor {PAF) receptor antagonists was measured by observing their inhibitory effects against PAF induced platelet aggregation. Their selectivity was assessed by monitoring their effect on platelet aggregation induced by arachidonic acid {AA) and adenosine diphosphate {ADP). The antagonists inhibited platelet aggregation induced at the PAF ECs0 {0.023 V_M)with the following rank order of potency: WEB 2086> WEB 2170> SRI 64-412> SRI 634]75> BN 52021> kadsurenone> SRI 63441> alprazolam. While the antagonists had no inhibitory effect at the ECs0 for A D P {10 t~I), they did inhibit platelet aggregation induced at the EC50 for AA {55 ttM). However, there was considerable variability in the slope of the inhibitory response and the relative potency of each antagonist against PAF induced platelet aggregation as compared to AA induced platelct aggregation. The antagonist ICs0 {p-M) against PAF and A A were as follows, with those that showed significantly different {p < 0.01) slopes indicated by an asterisk: SRI 63~141" {3.8, 15.1); SRI 6~675 {1.4, 36.2); SRI 64-412 {0.5, 10.5); BN 52021* {2.4, 58.9); kadsurenone* {2.8, 28.3); alprazolam* {10, 25}; WEB 2086 {0.055, 0.220), and WEB 2170 {0.107, 0.534). Therefore, in rabbit whole blood the antagonists were potent, although not completely selective, inhibitors of PAF induced platelet aggregation. These results suggest that the mode of action of PAF and AA induced platelet aggregation may share some common features. However, since the slope of the inhibitory response against PAF and AA for some antagonists differed, mechanistic differences in their action appear to exist. Lipids 26, 1189-1192 {1991).

effects at PAF ECs0 {i.e., the effective concentration required to induce 50% platelet aggregation}, and was expressed as the ICs0 {s the lowest concentration required to totally inhibit the EC50}. Antagonist selectivity was assessed by monitoring their effect on the platelet aggregation induced at the ECs0 for arachidonic acid {AA) and adenosine diphosphate {ADP}. MATERIALS AND METHODS

Whole blood (0.8volumes) was collected,using 20G teflon catheters (Surfl~ Terumo, Tokyo, Japan), from the marginal ear vein of male N e w Zealand White rabbits into 0.2 volumes of citrate 3.2% {w/v) {trisodium salt-dihydrate, Sigma Chemicals Co., St. Louis, MO). The citrated rabbit whole blood was added, in 50 ~L aliquots, to each of the eight wells of a teflon micromlxlng device, followed by 25 ~L of antagouist/vehicl~ After a 30-sec temperature equilibration period, 25 ~J~ of the agonist ECs0/vehicle was added. A 10/~L sample was taken immediately and after 15 mln incubation at 37~ for counting of the single, non-aggregated platelets (3-24/~rn3)using the Baker Instruments Series 810 Platelet Analyzer {Baker Instruments, Allentown, PA). The results were expressed as the platelet aggregation index, i.~,1-P1flP0,where P15 and P0 are the platelet count at 15 and 0 min, respectively. Therefore, an index of 0 signifiesno platelet aggregation and, conversely, 1 is total platelet aggregation. The platelet count for male N e w Zealand White rabbits was approximately 250 X 103/pL, and there was minimal variation between experimental animals. The bioassay was performed immediately after blood collection. P A F receptor antagonists examined were: SRI 63-441

{cis-{+ )-l-[2-[hydroxy[[tetrahydro-5-[(octadecylaminocarPlatelet-activating factor (PAF) induces platelet aggregation via receptor-mediated mechanisms. Using a rapid in vitro bioassay for the quantification of platelet aggregation (1), our aim was to measure the potency and selectivity of several PAF receptor antagonists in rabbit whole blood. The antagonists examined were either synthesized compounds based on the PAF structur~ such as SRI 63-441, SRI 63-675 and SRI 64-412; natural products, such as the terpene, BN 52021 and the neolignan, kadsurenone; or pharmacological agents such as the triazolobenzodiazepine alprazolam and the thienotriazolobenzodiazepines, WEB 2086 and WEB 2170. Antagonist potency was measured by observing their inhibitory 1Based on a paper presented at the Third International Conference on Platelet-Activating Factor and Structurally Related Alkyl Ether Lipids, Tokyo, Japan, May 1989. *To whom correspondence should be addressed. Abbreviations: AA, arachidonic acid; ADP, adenosine diphosphate; ECs0, activator concentration required to induce 50% platelet aggregation; IC5o,antagonist concentration required to inhibit platelet aggregation by the ECs0; PAF, platelet-activating factor; PBS, phosphate buffered saline.

bonyl)oxy]methyl] furan-2-yl]methoxy-phosphinyloxy]ethyl]quinolinium hydroxide, inner salt;Sandoz Research Institute, East Hanover, NJ) dissolved in calciurn/magnesinm-free phosphate buffered saline {PBS) {140 m M NaCI, 2.6 m M KCI, 8.1 m M Na2HPO4, 1.5 m M KH~P04; Sigma); SRI 63-675 {cis-{+)-l-[2-[hydroxy[[tetrahydrc~2, 5-dlmethyl-5-[{octadecylaminocarbonyl)oxy]methyl] furan2-yl]methoxy-phosphinyloxy]ethyl]quinolinium hydroxide, inner salt; Sandoz) dissolved in PBS; SRI 64-412 {5-[4'(3,4,5-trimethoxy phenylethyl)phenyl]-2,3-dihydroimidazo[2,1-a]isoquinolinehydrochloride; Sandoz) dissolved in 2% {v/v) acetic acid (AR grade; B D H Chemicals, Melbourne, Via, Australia) in distilled water; B N 52021 (9H-1,7a-(epoxymethano)-lH,6a-H-cyclopenta-[c]furo[2,3-b]furo[3',2':3,4]cyclopenta[1,2-d] furan-5,9,12-(4H)trione,3-tert-butylhexahydro-4,7b-ll-hydroxy-8-methyl; IHB-IPSEN, Le Plessis Robinson, France} dissolved in methanol {AR grade; BDH); kadsurenone (2-{3,4-dlmethoxyphenyl}-2,3-dihydro-3 a-methoxy-3-methyl-5-{aUyl)-62H-oxobenzohtran; Merck, Sharp and Dohme Research Laboratories, Rahway, N J) dissolved in ethanol {AR grade; BDH); alprazolam{8-chloro-l-methyl-6-phenyl-4H-2-triLIPIDS,Vol. 26, Na 12 0991)

1190

A.J. AMMIT AND C. O'NEILL azolo[4,3-a][1,4]benzodiazepine; Upjohn, Kalamazoa MI) dissolved in ethanol; WEB 2086 (3-[4-(2-chlorophenyl)9-methyl-6H-thieno[3,2-f] [1,2,4]triazolo-[4,3-a] [1,4]diazepin-2-yl]-l-(4-morpholinyl)-l-propanone; Boehringer Ingelheim KG, Ingelheim am Rhein, Germany) dissolved in PBS; and WEB 2170 (6-(2-chloro-phenyl)-8,9-dihydro-1methyl-8-(4-morpholinyl-carbonyl)-4H,7H-cyclopenta[4,5]thieno[3,2-f] [1,2,4]triazolo-[4,3-a] [1,4]diazepine; Boehringer Ingelheim) dissolved in PBS. All dilutions were in PBS. Activators used were: PAF (1-O-alkyl-2-acetyl-snglycero-3-phosphocholine; Sigma) dissolved in chloroform (AR grade; BDH), solvent was evaporated with N2 and the sample was resuspended and diluted in PBS containing 0.25% (w/v) bovine serum albumin (CSL, Melbourne, Vic., Australia); AA (from porcine liver; Sigma) dissolved in ethanol, solvent was evaporated with N2 and the sample was resuspended in 10 mM Na2CO3 (Sigma) and diluted in 0.9% (w/v) NaC1 (Sigma); and ADP (sodium salt, grade I, from equine muscle; Sigma) dissolved and diluted in 0.9% NaC1. The concentrations of activators used were the ECs0 as determined in a previous study (1). Statistical analysis was performed by estimating regression models for the linear section of the inhibition doseresponse curve (after correcting for differences in antagonist concentration), and by examining equality of slopes by analysis of covariance. RESULTS

PAF receptor antagonists inhibited platelet aggregation induced by PAF at the ECs0 {0.023 f~ml in rabbit whole blood with the following rank order of potency: WEB 2086> WEB 2170> SR164-412> SRI 63-675> BN 52021> kadsurenone> SRI 63-441> alprazolam. While the antagonists had no inhibitory effect at the ECs0 for ADP (10 ~M), they did inhibit platelet aggregation induced at the ECs0 for AA (55 ~!Vl). The ICs0 of the antagonist, the slope of the inhibitory response, and the relative potency of each antagonist against PAF and AA induced platelet aggregation are given in Table 1. Half of

TABLE 1 Effect of P A F Receptor Antagonists on P A F Induced and Arachidonic Acid Induced Platelet Aggregation in Rabbit Whole Blood a

Antagonist SRI 63-441b SRI 63-675 SRI 64-412 BN 52021b Kadsurenoneb Alprazolamb WEB 2086 WEB 2170

PAF AA Relative IC50 (~M) Slope IC50 (~M) Slope potency 3.8 1.4 0.5 2.4 2.8 10.0 0.055 0.107

--0.10 -0.28 -0.79 --0.18 -0.15 --0.03 --6.15 -3.64

15.1 36.2 10.5 58.9 28.3 25.0 0.220 0.534

-0.02 -0.28 -0.78 --0.06 -0.04 -0.09 -6.12 -2.96

4.0 25.8 21.0 24.5 10.1 2.5 4.0 5.0

aThe ICs0(~M),the slope of the inhibitoryresponse and the relative potencyof each PAF receptor antagonist against platelet aggregation inducedat the ECsofor PAF (0.023 ~_M)and AA (55 ~M) after 15 min incubationweredeterminedwithcitratedrabbit wholeblood. bSignificantlydifferent slopes at p < 0.01. LIPIDS,Vol, 26, No. 12 0991)

the antagonists examined showed significantly different (p < 0.01) slopes for the inhibition of PAF induced platelet aggregation as compared to AA induced platelet aggregation. Als~ the relative potency of each PAF receptor antagonist ranged widely, i.e., from 2.5 to 25.8. Based on these differences, the antagonists can be categorized into four groups: i) Those with parallel slopes and small differences in relative potency, i.e., WEB 2086 and WEB 2170; ii) those with different slopes and small differences in relative potency, i.e., SRI 63-441 and alprazolam; iii) those with parallel slopes and large differences in relative potency, i.e., SRI 63-675 and SRI 64-412; and iv) those with different slopes and large differences in relative potency, /.e., BN 52021 and kadsurenone The inhibitory responses for antagonists that are representative of each group are shown in Figure 1 (a-d). TO investigate whether the antagonists affected any of the whole blood cellular constituents or were acting as partial agonists, high concentrations of antagonists alone were incubated with rabbit whole blood and examined for blood smear abnormalities or platelet aggregation by phase contrast microscopy. WEB 2086 and WEB 2170 had no effect on the morphology of any of the whole blood cells when tested up to 55 ~M, while SRI 63-441 and SRI 63-675 caused platelet and erythrocyte lysis at 150 ~M and 362 ~M, respectively. None of these antagonists acted as partial agonists. There was difficulty, however, in performing the same examinations with antagonists dissolved in non-aqueous solvents, since the solvent alone can cause non-specific effects. The solvent for SR164-412 was 2% acetic acid in distilled water. At a concentration of 25000 ppm in citrated rabbit whole blood this solvent caused platelet and erythrocyte lysis (unpublished data). The solvent for BN 52021 was methanol, while kadsurenone and alprazolam were dissolved in ethanol. Methanol and ethanol at 100000 and 25000 ppm, respectively, completely blocked the ability of platelets to respond to an activating signal, since they non-specifically inhibited platelet aggregation induced by ADP, AA and PAF (unpublished data). Therefor~ due to the properties of the solvent, the highest antagonist concentrations that could be tested were 21, 118, 28 and 25 ~M for SRI 64-412, BN 52021, kadsurenone and alprazolam, respectively. At these concentrations the antagonists did not affect the morphology of the whole blood cellular constituents or induce platelet aggregation. Therefore, we can exclude the possibility that the results obtained in this study were due to the PAF-receptor antagonists acting as partial agonists or causing haemolysis of platelets or erythrocytes. DISCUSSION

In rabbit whole blood, P A F receptor antagonists were potent inhibitors of P A F induced platelet aggregation. The slope of the inhibition response was correlated with antagonist potency, and reflectsthe differingaffinitiesof the antagonists for the platelet P A F receptor (2).Comparative studies on the potency of P A F receptor antagonists are rare, and this is the firstwhich performs the examination on whole bloocL Since various experimental conditions can affect PAF-induced platelet aggregation, antagonist potency obtained from separate experiments cannot be strictly compared. Therefore, it is difficult to relate our

1191

PAF ANTAGONISTS AND ARACHIDONIC ACID EFFECTS 0.6

0.6

u~ zm

0.5

~--

0.4

~.~

0.4

r~q

0.3

~.~

0.3

I-

0-5 z

0.2

0-2 _1 w

uJ 0.1

.I.

0.0

q

00.005

f

!

I

!

0.010

0,022

0.055

0.110

I

0.1

i i

,&

0.0

0.220control

I

0 0"8

I

i

!

I

1.5

3.8

7.6

15,1

SRI 63-441 (I.OA)

WEe 2oee c~u)

0.6

0.6

o.5

i

control

0"5

(~

0.4

0.4

0-3

0.3

0.2

-i tu

!

0.2

0.1

0.1

i i

o.o

~ ! 0 0.4

I

I

I

!

0.7

1.4

7.2

14'5

SR! 6 3 - 6 7 5

I 36"2 (ontroI

(J~M)

0"0

&

I

0 0'6

|

I

I

I

I

I

1-2

2.4

5.9

11"8

23"6

58'9

l

s

ON 5 2 0 2 1 (/xM)

FIG. 1. Effect of increasing concentrations of PAF receptor antagonists on platelet aggregation induced at the ECs0 for PAF (0.023 pM; I ) and A A (55 pM; e), after 15 min incubation with citrated rabbit whole blood. Each point represents the mean ___SEM of >~3 replicates. The effect of vehicle alone is represented by the histogram (control), which is the mean ----- SEM of >/11 replicates.

results to published values. However, some of the antagonists used in this study were examined in a comparative study with human platelet-rich plasma using ttm bidometric aggregometry (3). These authors obtained a similar rank order of potency, WEB 2086> BN 52021> kadsurenone> alprazolan~ The PAF receptor antagonists were not completely selective for PAF since, at concentrations that were 2.5to 25.8-fold higher than required for the inhibition of PAF, they also inhibited platelet aggregation induced by AA. For four of the eight PAF receptor antagonists studied, i.e., SRI 63-675, SRI 64-412, BN 52021 and kadsurenone, the antagonist concentration required to inhibit AA induced platelet aggregation were 10-fold higher than those required for PAF. Although this may be considered reasonably selective by pharmacological standards, these results are in contrast to previous reports where PAF receptor antagonists were shown to have either no effect on AA induced platelet aggregation (4,5}, or only minimal effects requiring concentrations that were 200- to ll00fold higher than those required for inhibition of PAF {6,7}. These studies were performed using isolated human platelets. There are a number of possible explanations for the inhibition of AA induced platelet aggregation by PAF receptor antagonists, although we are not able to define the

actual mechanism from this study. The effect could be due to either inhibition of the (membrane~associated) enzymes responsible for the conversion of AA to its activator products, prostaglandin G2, prostaglandin H2 and thromboxane A2; or via direct competitive inhibition of the platelet receptor for these AA metabolites. Currently there is no literature supporting these suggestions. A more likely explanation, given the mixed population of cells in whole blood and the increasing evidence supporting the involvement of AA in PAF induced platelet aggregation (8,9), is that PAF may be involved in AA induced platelet aggregation. This could be either autocrine (/.e., AA induces platelets to produce PAF) or paracrine (i.e., generation of PAF by other cells in the whole blood, e.g., neutrophils). Although there was a general trend for the slope of the in~bitory response against AA induced platelet aggregation to be similar to that against PAF in magnitude, half of the antagonists showed significant differences. This, together with the wide range of relative potency observed, suggests that the inhibitory effect of the antagonists on AA induced platelet aggregation is not solely via a direct competitive inhibition of the platelet PAF receptor. Therefore, the PAF receptor antagonists may be demonstrating intrinsic pharmacological selectivity for the PAF receptor, as previously demonstrated in defined systems which examined isolated cell types, but LIPIDS, Vol. 26, No. 12 (1991)

1192

A.J. AMMIT AND C. O'NEILL the results are a reflection of the mixed population of cells in whole blood and the poorly understood complex interplay understood complex interplay between PAF and A A metabolites. Thereforc~ in an in vitro bioassay which uses whole blood to closely mlm]c the in v i v o situation, PAF receptor ant a g o n i s t s were potent, although not completely selective inhibitors of PAF induced platelet a g g r e g a t i o n To clarify the nature of the inhibition of A A induced platelet aggregation b y PAF antagonists, and to allow a b e t t e r comparison with previous data, it would be of interest to perform these experiments with rabbit platelets prepared as platelet-rich p l a s m a and]or washed platelets, and to examine h u m a n platelet responses as well Our results would a p p e a r to be relevant to in v i v o studies using PAF receptor antagonists and indicate t h a t whole blood seems more appropriate for the in v i t r o a s s e s s m e n t of a n t a g o n i s t selectivity.

ACKNOWLEDGMENTS The authors would like to thank Michael P. Jones for his assistance with the statistical analysis and the following companies for their generous gifts of PAF receptor antagonists: Sandoz Research

LIPIDS,Vol. 26, Na 12 (1991)

Institute; IHB-IPSEN; Merck, Sharp and Dohme Research Laboratories; Upjolm; and Boehringer Ingelhein~

REFERENCES 1. Ammit, A.J., and O'Neill, C. {1991) J. Pharmacot Methods 26, 7-21. 2. Braquet, P., Touqni, L, Shen, TY., and Vargaftig, B.B. (1987}PharmacoL Rev. 39, 97-145. 3. Heuer, H.H., Casals-Stenzel, T., and Weber, K:H. (1987} in Proceedings of the lOth International Congress of Pharmacology Satellite Meeting on Recent Advances in Platelet-Activating Fac-

tor, IX 16, Australia 4. Kornecki, E., Ehrlich, Y.H., and Lenox, R.H. (1984} Science 226, 1454-1456. 5. Chesney, C.M., Pifer, D.D., and Cagen, L.M. {1987} Biachem. Biophys. Res. Commun. 144, 359-366. 6. Ntmez, D., Chignard, M., Korth, R., Le Couedic, J-P., Norel, X., Spinnewyn, I3.,Braquet, P., and Benveniste, J. (1986) Eur. J. Phar~ macoL 123, 197-205. 7. Casals-Stenzel, J., Muacevic, G., and Weber, K.-H. {1987)J. Phar~ macoL Exp. Therap. 241, 974-981. 8. Macconi, D., Morzenti, G., Livic~M., Morelli, C., Cassina, G., and Remuzzi, G. (1985) Lab. Invest. 52, 159-168. 9. Sturk, A., Schaap, M.C.L., ten Cate, J,W., Heymans, H.S.A., Schutgens, R.B.H., PrzyrembeL H., and Borst, P. (1987) J. Clin. Invest. 79, 344-350. [Received September 6, 1989, and in revised form October 3, 1991; Revision accepted October 7, 1991]