Camp. Biochem. Physiol. Vol. 99A, No. 4, pp. 551-557,

0300-9629/91

1991

$3.00 + 0.00

0 1991Pergamon Press plc

Printed in Great Britain

EFFECT OF CALCIUM CONCENTRATION AND INHIBITORS ON THE RESPONSES OF PLATELETS STIMULATED WITH COLLAGEN: CONTRAST BETWEEN HUMAN AND RABBIT PLATELETS M.

A. PACKHAM,*

M. L.

RAND,

D. H.

RUBEN

and R. L. KINLOUGH-RATHBONEf

Department of Biochemistry, University of Toronto, Toronto, Ontario, Canada, M5S 1A8, Telephone: (416) 978-8567; TDepartment of Pathology, McMaster University, Hamilton, Ontario, Canada (Received 16 October

1990)

Abstract-l. Variations in the concentration of Ca2+ [Ca2+] in the suspending medium have different effects on the responses of human and rabbit platelets to collagen. 2. When rabbit platelets are stimulated with a low concentration of collagen (0.5 pg/ml), aggregation, release of granule contents, and formation of thromboxane are maximal when the suspending medium contains [Ca2+] in the physiological range (0.5-2.0 mM), and very slight in a medium with no added Ca2+. 3. In contrast, human platelets respond most strongly when the suspending medium contains no added Ca2+ ([Ca2+] approx. 20 PM); this is attributable to the enhanced formation of thromboxane A, (TXA>) upon close platelet-to-platelet contact in this medium. 4. When TXA2 formation is blocked by inhibition of cycle-oxygenase with aspirin or indomethacin, rabbit platelet aggregation and release in response to 1.25-10 pg/ml collagen is also maximal at [Ca*+] of 0.5-2.0 mM and least at 20 PM; human platelets do not aggregate and the extent of release is relatively independent of [Ca2+]. 5. In 1 mM [Ca2+], use of apyrase and/or ketanserin with rabbit platelets in which TXA, formation is blocked shows that released ADP and serotonin make large contributions to aggregation and release in response to high concentrations of collagen; human platelet aggregation is largely dependent on TXA,. 6. Use of fura-2-loaded platelets shows that the collagen-induced rise in cytosolic [Ca2+] is only slightly inhibited by aspirin or indomethacin in rabbit platelets, but almost completely inhibited in human

platelets. 7. Responses of rabbit platelets to collagen are less dependent on TXA, than those of human platelets. Released ADP and serotonin make major contributions to the responses of rabbit platelets to collagen.

INTRODUCTION

Earlier studies indicated that during collagen-induced aggregation the synergistic effect of thromboxane A, (TXA,) formed by rabbit or human platelets adherent to collagen and of the ADP they released was responsible for the aggregation of platelets, including those that were not adherent to collagen (KinloughRathbone et al., 1977, 1980; Packham et al., 1977). We also showed previously that neither aspirin nor indomethacin inhibited the release of dense granule contents from rabbit platelets adherent to a collagencoated surface, and that these drugs had little or no inhibitory effect on adhesion (Kinlough-Rathbone ef al., 1980). EDTA prevented release of granule contents from adherent platelets, but also greatly reduced adhesion (Kinlough-Rathbone et al., 1980). In another study, higher concentrations of prostaglandin E, (PGE,) or prostaglandin I, (PGIr) were required to inhibit release of granule contents than were required to inhibit adherence of rabbit platelets to collagen-coated glass or the subendothelium, but results with human platelets differed in that PGE, and PG12 at concentrations as high as 10 PM failed to inhibit release of dense granule contents from

*Author to whom correspondence should be addressed.

platelets adherent to a collagen-coated surface, although platelet adhesion was reduced to about one-half (Cazenave et al., 1979). When we were investigating the effects on platelet responses of various concentrations of Ca2+ in the platelet suspending medium (Packham et al., 1989) differences between the responses of rabbit and human platelets to collagen became apparent. We have now extended these studies of the effect of the concentration of Ca2+ in the medium to an examination of platelet responses to high concentrations of collagen when the formation of thromboxane A, is blocked with aspirin or indomethacin. Since we observed considerable aggregation of rabbit platelets when the activity of cycle-oxygenase was blocked, and noted appreciable release of granule contents by both rabbit and human platelets in this condition, we investigated the role of released dense granule contents in these responses. We also determined the effect of inhibitors of cycle-oxygenase on the mobilization of internal Ca*+ in response to collagen with platelets from both species. MATERIALS

AND METHODS

Materials were obtained from the following suppliers: bovine tendon collagen, acetylsalicylic acid (aspirin), indomethacin, imipramine, creatine phosphate (CP), creatine 551

M. A. PACKHAM et al.

552

phosphokinase (CPK), ethylene glycol-bis(p-aminoethyl ether)N,N,N’,N’-tetraacetic acid (EGTA) and N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (Hepes) from the Sigma Chemical Co. (St. Louis, MO, USA:); ethylenediaminetetraacetic acid. disodium salt (EDTA) from BDH Chemicals (Toronto, bntario, Canada); ketanserin, potassium tartrate salt, from Janssen Pharmaceutics (Mississauga, Ontario, Canada): albumin (fraction V) from ICN ImmunoBiologicals (Lisle, IL); prostaglandin E, (PGE,) from Upjohn (Kalamazoo, MI); furaacetoxymethyl ester (furaAM) from Molecular Probes (Eugene, OR); 14Cserotonin (S-hydroxytryptamine-3’-Y-creatinine sulfate, 60 mCi/mmol) from Amersham Corp. (Oakville, Ontario, Canada); and radioimmunoassay kits for thromboxane B, (TXB,) from NEN (Lachine, Quebec, Canada). Acetic acid soluble collagen was prepared as previously described (Cazenave ef al., 1973) stored at 4°C and diluted with 0.15 M NaCl before use. Apyrase was prepared by the method of Molnar and Lorand (1961) and used as described elsewhere (Kinlough-Rathbone et al., 1983) (2 PI/ml of platelet suspension) or, as indicated, in higher concentrations (lOOpl/ml of platelet suspension) in some experiments All concentrations are given as final concentrations after all additions to the platelet suspensions. Suspensions of washed platelets from human subjects or from rabbits (New Zealand White) were prepared as previously described (Kinlough-Rathbone et al., 1983; Ardlie et al., 1970, 1971). Human subjects had not taken medications that affect platelet function for at least 2 weeks prior to donation of blood. The final suspending medium for platelets from both species was a modified Tyrode solution from which calcium salts were omitted; it contained 1 mM MgCl,, 137 mM NaCl, 2.68 mM KCI, Il.9 mM NaHCO,, 0.42 mM NaH,PO, H,O, 5.5 mM glucose, apyrase (2 PI/ml) and 0.35% albumin (pH 7.35). The concentration of Ca*+ in the suspending medium of the platelets was approximately 20pM as measured by atomic absorption spectrophotometry. Human platelets were stored at 37°C rabbit platelets were stored at room temperature and warmed to 37°C for 30 min before use. The platelet counts were adjusted to 500,000/~1. To permit measurement of the extent of the release of the contents of the dense granules, the platelets were labelled in the first washing solution with “C-serotonin (I PCijlO ml of washing fluid), and release of “C-serotonin was determined as described previously (Greenberg et al., 1975). Imipramine

&

RelZase TX& 50 4.3 50 45

5.7

6.5

91 97 100 80

26

4.0

Fig. I. Effect of the concentration of added Ca’+ in the suspending medium of rabbit platelets on aggregation, per cent release of i4C-serotonin, and TXB, formation (ng/109 platelets) induced by acid soluble collagen (COLL, 0.5 pgg/ml). Typical of two experiments.

(5pM, final concentration) was added before collagen to prevent the reuptake of released serotonin. Platelet aggregation was recorded at 37°C in an aggregation module (Payton Associates, Scarborough, Ont., Canada) at a stirring speed of 1,100 rpm. TXB, was measured by radioimmunoassay in supernatant samples prepared by centrifugation of platelet suspensions for 1 min at 12,000 g in an Eppendorf centrifuge (Brinkman, Rexdale, Ont., Canada); results were calculated as ng/109 platelets. In experiments in which the activity of cycle-oxygenase was inhibited, platelets were treated with aspirin (500 pM) in the first washing solution; alternatively. indomethacin (10 PM) was added to the platelet suspending medium 5 min before the addition of collagen. Other inhibitors were also added 5 min before collagen. In some experiments, platelets in the-first washing solution (Ca2+-free Tvrode-albumin with 5.3 mM EGTA) at a platelet count of SdO,OOO/~l, were loaded with fura- AM by the method of Pollock ef al. (1986) by incubation at 37 C for 15 min. After the usual washing procedure, the platelets were resuspended in a modified Tyrode-albumin solution (no added Ca2+, 5 mM Hepes), platelet count 2OO,OOO/~l. Fluorescence measurements were made with a SPEX Fluorolog 2 Spectrophotometer (SPEX Industries, Metuchan. NJ, U.S.A.) with excitation at 339nm and emission at 500 nm. RESULTS

Rabbit platelets

As shown previously (Packham et al., 1989) rabbit platelet responses to a low concentration of collagen (0.5 pg/ml) were strongly inhibited by a lack of Car+ in the suspending medium (Fig. 1). Aggregation, release of “C-serotonin and formation of thromboxane were greatest when the concentration of Ca’+ was between 0.5 and 2.0 mM. Aggregation did not occur and release and thromboxane formation were very slight in a medium with no added Car+. When rabbit platelets were treated with aspirin or indomethacin (at concentrations that block collageninduced TXB, formation and inhibit aggregation caused by arachidonic acid), aggregation and release of granule contents in response to higher concentrations of collagen (1.25 to 10 pg/ml) were also of 0.5 to 2.0 mM in greatest at Ca2+ concentrations the suspending medium and much less with lower concentrations of Ca2+ or with 5 mM Ca” (Fig. 2 and Table 1). EDTA or EGTA blocked aggregation and reduced the release of granule contents to low values (3-6%), but the platelets changed shape in response to collagen. Apyrase (100 PI/ml) strongly inhibited aggregation of indomethacinor aspirin-treated platelets suspended in a medium with 1 mM Ca2+ (Fig. 3) and reduced the release of “C-serotonin by these platelets to 54.7 If: 13.5% (mean If: SD) of the value without apyrase; ketanserin (100 PM) also strongly inhibited aggregation and reduced release to 65.4 + 9.3% of the value without ketanserin; the combination of apyrase and ketanserin reduced the release of granule contents to even lower levels (40.9 _t 8.45% of the value without apyrase or ketanserin) (n = 5 in all cases). Similar results were obtained when CPjCPK (4 PM/~ Units/ml) was used instead of apyrase to remove released ADP (data not shown). When the platelets were not treated with an inhibitor of cycle-oxygenase, only partial inhibition of release of “C-serotonin and formation of thromboxane was observed in the presence of apyrase.

Rabbit/human

platelet

responses

553

to collagen

Ca2+ X ImMl --- Release TX& 1.0

F&ate

626

60.6

COlltrOl

Control

Apyrare Ketantsrin pyrsrin pyrars

TX82

65.5 57.7 54.6

568 636 469

49.6

463

1 Not Indo. Treated J

1 min.

Control

25.2

0.5

1 I

0.6

5.0

26.6

0.1

19.3

0.6

C8.O” ‘9:$

0.5 0.4

Indo. Treated

Indo. Treated

0.5 1z 1012

Fig. 2. Effect of the concentration of Ca2+ in the suspending medium of rabbit platelets on aggregation, per cent release of “‘C-serotonin and formation of TXB, (ng/109 platelets) induced by a high concentration of collagen (COLL, 5 pg/ml) in the presence of indomethacin (10pM) added

5 min before collagen. Typical of six experiments with indomethacin and three experiments with aspirin-pretreated platelets. ketanserin, or the combination of these inhibitors, upon stimulation of the platelets with 5 pg/ml of collagen (Fig. 3). With indomethacin-treated platelets, 10 PM PGE, reduced the extent of collagen-induced (5 @g/ml) release of “C-serotonin from 32.8 to 5.6% at 1 mM Ca*+ and from 8.7 to 3.7% in the medium without added Ca’+. In the presence of EGTA, PGE, reduced release of “C-serotonin from indomethacin-treated platelets from 5.0% to 2.0%. Stimulation of rabbit platelets with collagen (1.25 pg/ml) in a medium containing 1 mM Ca2+ caused a rise in internal Ca*+ after a lag phase, and aggregation eventually occurred; this increase in cytosolic Cal+ was only slightly inhibited by aspirin, and inhibition was not apparent during the initial collaTable

I. Effect of the concentration

medium

on the release

stimulated

with

of Ca2+

of 14C-serotonin

collagen

after

added to the suspending

from

treatment

prelabelled

with

platelets

aspirin

or

indo-

methacin Release (%

at I mM

(mean

ofCa’+

Rabbit 22.7 + 9.2

91.2k

0.01

29.8 k 5.9

90.6 f 9.8

0.05

41.3 f

95.1*

0.10

64.4 k 6.0

0.50

98.9 k 10.5

12.6

I oat 99.9 + 1.9

5.00

77.8 +_ 8.7

94.0 +_ 2.3

collagen

10.3 to 69.5%

ASA-pretreated

100.0 * 4.5

96.6 & 6.1

from

$O--

n=6 of

“C-serotonin

over the range

at

1 mM

of concentrations

Ca2+ of

used (I .25- IO fig/ml).

human

platelets,

16.4 to 29.0% (S--lO~gg;ml).

over

the

release

the range

Control

96.0 2 2.5

2.00

the release

-

3.5

100’

varied iWith

3.9

I .oo

platelets,

As shown previously (Packham er al., 1989), with low concentrations of collagen, aggregation, release, and TXB, formation were greatest when no Ca*+ had been added to the medium in which human platelets were suspended. (The concentration of Ca2’ in this medium is approximately 20 PM.) Platelet responses at 1.0 mM Ca2+ were greater than at 0.5 mM or 2.0 mM (Fig. 5). When human platelets were treated with aspirin or indomethacin to block TXA, formation, high concentrations of collagen caused only a slight increase in light transmission at any concentration of Ca*+ in the medium (Fig. 6), and for a given concentration of collagen, the extent of release of “C-serotonin was very little different at all concentrations of added Ca2+ from 0 to 5 mM (Table 1).

Human

0

rabbit

Human platelets

z41

n = II *With

gen-induced rise in the concentration of cytosolic Ca2+ (Fig. 4). When the platelets had been treated with aspirin, aggregation did not occur in this slowly stirred system.

Ca*+)

k SD)

-

(mM) ~______

Fig. 3. Effect of ketanserin (100 PM), 100 $/ml of apyrase, and the combination of these inhibitors on aggregation, per cent release of 14C-serotonin and formation of TXBz (ng/109 platelets) by rabbit platelets induced by a high concentration of collagen (COLL, 5 pg/ml) in the presence and absence of indomethacin (10 PM). (The suspending medium contained 1 mM Ca2+.) Mean values from three experiments are shown.

of “C-serotonin

of release

Concentration

:::

of

“C-serotonin

of collagen

varied

concentrations

from used

TIME

(min.1

Fig. 4. Increase in cytosolic Cal+ in fura-2-loaded rabbit platelets stirred with collagen (COLL, 1.25 /cg/ml) at 37°C. (A) control platelets; (B) aspirin- (ASA-)pretreated platelets. (The suspending medium contained 1 mM Ca2+.)

M. A. PACKHAM et al.

554

109 116 11s 17

58

11

50

93

46

6.1 2.7 1.9

:; 36

I

Fig. 5. Effect of the concentration of added Cal+ m the suspending medium of human platelets on aggregation, per cent release of ‘%-serotonin, and TXB, formation (ng/109 platelets) induced by acid soluble collagen (COLL). (A) Collagen 0.25 ng/ml. (B) Collagen 0.4 pggiml. Typical of four similar experiments.

With high concentrations of collagen in a medium containing 1 mM Ca*+, addition of ketanserin, or a high concentration of apyrase, or the combination of these inhibitors only partially inhibited aggregation, release of granule contents and formation of thromboxane (Fig. 7). In the presence of indomethacin, the greatly reduced extent of release was further diminished by the presence of ketanserin, apyrase, or both (Fig. 7). EDTA, EGTA, or PGE, strongly inhibited aggregation in response to a high concentration of collagen (5 pg/ml), but had less effect on the release of dense

1min.

f

-Ca2+ % lmM)Release .TXB2 1.0

72.6

710

Control

granule contents (76.2 k 3.6% in the presence of 1 mM Ca2+, 34.6 + 4.6% in the presence of 4mM EDTA, 46.9 k 4.7% in the presence of 5 mM EGTA. and 20.1 + 2.7% in the presence of 10 PM PGE,). When indomethacin was present, the reduced extent of release was further diminished by the presence of each of these inhibitors (14.0 + 3.3% in the medium with 1 mM Ca’+, 11.6 k 1.O% with 4 mM EDTA, 9.0 + 1.5% with 5 mM EGTA, and 10.2 + 0.5% with 10 PM PGE,). In keeping with the observations of other investigators (Pollock et al., 1986) we found that aspirin pretreatment of the platelets or the presence of indomethacin almost completely blocked the delayed rise in internal Cal+ in response to a low concentration of collagen (Fig. 8).

F&e TXB, Control Ketanserr

61.3 80.0

827 569

65.3

644

73.4

761

1

Not Indo.

Treated

1 min.

In&. Tyated

Fig. 6. Effect of the concentration of added Ca’+ in the suspending medium of human platelets on aggregation, per cent release of “C-serotonin and formation of TXB, (ng/109 platelets) induced by a high concentration of collagen (COLL, 5 pg/ml) in the presence of indomethacin (lOpM), added 5 min before collagen. Typical of three experiments with indomethacin and two experiments with aspirin-pretreated platelets.

1

13.0 12.2

0.9 0.9

Indo.

/@

‘$3

Tyted

Fig. 7. Effect of ketanserin (100 PM), apyrase (100 pljml). and a combination of these inhibitors on aggregation, per cent release of “C-serotonin and formation of TXB, (ng/109 platelets) from human platelets induced by a high concentration of collagen (COLL, 5 pg/ml) in the presence and absence of indomethacin (10 PM). (The suspending medium contained 1 mM Ca2+.)

Rabbit/human

platelet

Fig. 8. Increase in cytosolic Ca2+ in fura-24oaded human olatelets stirred with collaeen (COLL, 1.25,_, ugiml) , at 37°C. control platelets; (B) aspirin- (ASA-)pretreated platelets. (The suspending medium contained 1 mM Ca2+.)

iA)

DlSClJSSlON

These studies were initiated because so many investigations of platelet reactions have been done in titrated platelet-rich plasma or in artificial media to which no calcium salts have been added; we have found major differences between platelet reactions in these low Ca*+ media and reactions in the presence of physiological concentrations of Ca2+ (Packham et al., 1989) and we wish to underscore the point that reactions in low Ca*+ media do not reflect the responses that would be likely to occur in an in uivo situation. In addition, some of the differences between the reactions of rabbit and human platelets are less evident when the concentration of Ca2+ in the suspending medium is in the physiological range. For example, at a physiological concentration of Ca’+, both human and rabbit platelets undergo only primary, reversible aggregation in response to ADP; however, in a medium with a micromolar concentration of Ca*+, ADP induces two phases of aggregation of human platelets, the second phase being accompanied by formation of thromboxane A, and release of granule contents (Packham et al., 1987, 1989). In contrast, in a low Ca2+ medium, rabbit platelets undergo only the primary, reversible phase of aggregation in response to ADP. The reason is unknown for the activation of phospholipase A* and the subsequent formation of TXA, in human platelets, but not in rabbit platelets, upon close platelet-to-platelet contact in a low Ca2+ medium. Effect of the concentration medium

qfCa*’

in the suspending

As shown previously (Packham et al., 1989) when rabbit platelets are stimulated with low concentrations of collagen, aggregation, release of granule contents and formation of thromboxane are at a maximum when the suspending medium contains Ca*+ at concentrations in the physiological range (between 0.5 and 2.0 mM) and very slight in a medium with no added Ca*+ (concentration of Ca*+ approximately 20 p M). In contrast to these results with rabbit platelets, low concentrations of collagen (that caused a maximum release of less than 30% of “C-serotonin) induced the strongest responses of human platelets when the suspending medium contained no added Ca*+. The stronger responses of human platelets in a medium with micromolar concentrations of Ca*+ are undoubtedly attributable to

responses

to collagen

555

augmentation of the amount of TXA, formed upon close platelet-to-platelet contact in this medium (Packham et al., 1987). Unlike human platelets, rabbit platelets do not form TXA, upon close plateletto-platelet contact in a low Ca2+ medium (Packham et al., 1989) and therefore the responses of rabbit platelets to collagen are not enhanced in this medium. We also observed that when rabbit platelets were treated with aspirin or indomethacin, the concentration of Ca2+ in the suspending medium had the same influence on the responses to a high concentration of collagen as it had on the responses to a low concentration of collagen by platelets in which cycle-oxygenase had not been inhibited. That is, aggregation and release of dense granule contents were greatest when the concentration of Ca’+ was in the physiological range, and these responses were largely abolished when Ca2+ was omitted from the suspending medium. In contrast to rabbit platelets, when human platelets were treated with aspirin or indomethacin to block the formation of TXA, and stirred with high concentrations of collagen, very little aggregation occurred at any concentration of Ca*+ in the medium and although the extent of release was greatly reduced, it was relatively independent of the concentration of added Ca2+ over the range of 0 to 5 mM. These observations indicate that collagen-induced aggregation of human platelets is largely dependent on TXA, at all concentrations of Ca*+ and that when TXA, formation is blocked, released ADP and serotonin are ineffective in causing aggregation, even when 20 to 25% of the dense granule contents have been released. Based on the value of releasable ADP of 2.5 ~mol/lO” platelets given by Lages (1986), the concentration of released ADP should have been 2.5 PM which, if added directly in the presence of fibrinogen would be sufficient to cause extensive aggregation (Mustard et al., 1972). However, assuming that a similar extent of release of the contents of the alpha granules occurred, the concentration of fibrinogen in the suspending medium would have been only about 30pg/ml (Cattaneo et al., 1987) which is considerably less than the optimum for ADP-induced aggregation (Kinlough-Rathbone et al., 1983). Alternatively, it may be that the release of ADP is sufficiently gradual that the platelets become refractory (Born and Cross, 1963; O’Brien, 1966; Holme and Holmsen, 1975; Hallam et al., 1982) before the concentration builds up to an effective level. Since the concentration of Ca2+ in the medium of human platelets in which cycle-oxygenase had been inhibited did not influence the extent of release of dense granule contents, and it is likely that with the high concentration of collagen the materials that were released originated from the platelets actually adherent to collagen (ADP does not cause release of granule contents from human platelets when TXA, formation is prevented, Packham et al., 1989), it seems reasonable to conclude that the extent of release from these adherent platelets is not dependent on the concentration of Ca’+. The reaction of the adherent platelets can be described as “directly induced platelet activation” as defined by Charo et al. (1977); that is, secretion occurs in the absence of aggregation and is not blocked by indomethacin.

556 Effect

M. A. PACKHAM ei al. of combinations of inhibitors at 1 mM

REFERENCES

Ca*+

In our previous studies (Kinlough-Rathbone er al., 1977, 1980; Packham et al., 1977) we showed that platelet aggregation in response to low concentrations of collagen was dependent on the synergism between TXA, and ADP, and that aggregation could be inhibited either by preventing TXA, formation or removing released ADP; the extent of inhibition was dependent on the concentration of collagen. When we used high concentrations of collagen, aggregation of rabbit platelets was only slightly inhibited by indomethacin or the creatine phosphate/creatine phosphokinase system which converts released ADP to ATP; the combination of these inhibitors further reduced the release of dense granule contents, but did not abolish it (Kinlough-Rathbone et al., 1980). In the present study, we have confirmed these observations and shown that in addition to released ADP, released serotonin also contributes to the responses of rabbit platelets because ketanserin has an additional inhibitory effect when cyclooxygenase activity has been blocked and ADP has been removed with apyrase. However, even when all these inhibitors are present, although aggregation is blocked, a slight release of granule contents occurs from platelets that we assume are adherent to collagen. With platelets from both species, EDTA, EGTA or PGE, inhibited aggregation and the release of granule contents in either the absence or presence of indomethacin. Thus external Ca*+ is required for these responses (although with human platelets, concentrations as low as 20pM are sufficient) and these responses are inhibited when the concentration of cyclic AMP is raised. The finding that the collagen-induced rise in cytosolic Ca*+ in human platelets is almost completely inhibited by aspirin or indomethacin is in accord with the observations of other jnvestigators who studied human platelets (Pollock et al., 1986) and rat platelets (Nakano et al., 1987). However, with rabbit platelets, the collagen-induced rise in cytosolic Ca2’ is only slightly inhibited by blocking the activity of cyclooxygenase. These observations may be related to the extensive aggregation and release seen with aspirinor indomethacin-treated rabbit platelets that does not occur with similarly treated human platelets in response to a high concentration of collagen. It may be that responses of rabbit platelets to collagen are largely mediated by released ADP and serotonin which together stimulate the rise in cytosolic Ca*+ even when TXA, formation is blocked, but the responses of human or rat platelets to collagen are more dependent on TXA, and when its formation is blocked, insufficient ADP and serotonin are released to produce a rise in cytosolic Ca*+. Species differences have been reported in the extent of the contribution of thromboxane A, to collagen-induced aggregation (Mallarkey and Smith, 1985). AcknoM,(ednements-The excellent technical assistance of Miss Nancy L. Bryant is greatly appreciated. This study was sunoorted by a grant. MT 2629, from the Medical Researdi Council of Canada. M. L. Rand is a Research Scholar supported by the Heart and Stroke Foundation of Ontario.

Ardlie N. G., Packham M. A. and Mustard J. F. 11970) Adenosine diphosphate-induced platelet aggregation in suspensions of washed rabbit platelets. Br. J. Ifwma/of. 19, 7-17. Ardlie N. G., Perry D. W.. Packham M. A. and Mustard J. F. (1971) Iniluence of aovrase on stability of washed rabbit‘ plaielets. Proc. &k. E.Y~. Biof. . Med. 136, 1021-1023. Born G. V. R. and Cross M. J. (1963) The aggregation 01 blood platelets. J. Physiof. 168, 178..195. Cattaneo-M., Kinlough-Rathbone R. L., Lecchi A., Bevilacaua C.. Packham M. A. and Mustard J. F. (1987) Fibrinogen-independent aggregation and deaggregation of human platelets: studies in two afibrinogenemic patients. Nood 70, 22 I-226. Cazenave J.-P., Dejdna E., Kinl~u8h-Rathbone R. L.. Richardson M., Packham M. A. and Mustard J. F. (1979) Prostaglandins I? and E, reduce rabbit and human platelet adherence without inhibiting serotonin release from adherent platelets. Thromb. R& 15, 213-279. Cazenave J.-P.. Packham M. A. and Mustard J. F. (1973) Adherence plateletsto a collagen-coated surface: development of a quantitative method. J. Lab. C/in. Med. 82, 978-990. Charo I. F., Feinman T. C. and Detwiler T. C. (1977) Interrelations of platelet aggregation and secretion.

of

J. Clin. hvest. 60, 866-873.

Greenberg J., Packham M. A., Cazenave J.-P.. Reimcrs H.-J. and Mustard J. F. (1975) Effects on platelet function of removal of platelet sialic acid by neuraminidase. Lab. Invesr. 32, 476-484. Hallam T. J., Ruggles P. A., Scrutton M. C. and Wallis R. B. (1982) Desensitisation in human and rabbit blood platelets. Thromb. Haemosl. 41, 278-284. Holme S. and Holmsen H. (1975) ADP-induced refractory state of platelets in r:ituo I. Methodological studies on aggregation in platelet rich plasma. Stand. J. Huematoi. 15, 96-103.

Kinlough-Rathbone R. L., Cazenave J.-P., Packham M. A. and Mustard J. F. (1980) Effect of inhibitors of the arachido~ate pathway on the release of granule contents from rabbit platelets adherent to collagen. Lab. Inwst. 42. 28-34.

Kinlough-Rathbone R. L.. Packham M. A. and Mustard J. F. (1983) In Methods in Hemafology. Measuremenrs of Piafeler Function (Edited by Harker L. A. and Zimmerman T. S.), pp. 64-91. Churchill Livingstone, Edinburgh. Kinlough-Rathbone R. L., Packham M. A., Reimers H.-J.. Cazenave J.-P. and Mustard J. F. (1977) Mechanisms of platelet shape change, aggregation and release induced by collagen. thrombin or A23.187. J. Lah. Cfin. Med. 90, 707-719.

Lages B. (1986) Studies on storage mechanisms in the dense granules: Storage pool deficient platelets. In Platelet Responses and Metabolism (Edited by Holmsen H.), Vol. II, pp. 13%-151. CRC Press, Boca Raton. FL. U.S.A. Mallarkey G. and Smith G. M. (1985) A comparative study of the involvement of the prostaglandin H,/thromboxane A, pathway in intravascular platelet aggregation in guinea pigs and rats. Br. J. Pharmacol. 84, 425430 Molnar J. and Lorand L. (1961) Studies on apyrases. .&rit. Biochem. Biophys. 93, 353-363.

Mustard J. F., Perry D. W., Ardlie N. G. and Packham M. A. (1972) Preparation of suspension of washed platelets from humans. Br. J. Haematol. 22, 193-204. Nakano T., Hanasaki K. and Arita H. (1987) Involvement of the Na+/H+ antiporter in activation of rat platelets by collagen. J. Biochem. 102, 677-680. O’Brien J. R. (1966) Changes in platelet membranes possibly associated with platelet stickiness. Nature 212, 1057-1058.

Rabbit/human

platelet

Packham M. A., Bryant N. L., Guccione M. A., KinloughRathbone R. L. and Mustard J. F. (1989) Effect of the concentration of Car+ in the suspending medium on the responses of human and rabbit platelets to aggregating agents. Thromb. Haemost. 62, 968-976. Packham M. A., Kinlough-Rathbone R. L. and Mustard J. F. (1987) Thromboxane A, causes feedback amplification involving extensive thromboxane A, formation upon close contact of human platelets in media with a low concentration of ionized calcium. Blood 70, 64765 1.

responses

to collagen

551

Packham M. A., Kinlough-Rathbone R. L., Reimers H.-J., Scott S. and Mustard J. F. (!977) Mechanisms of platelet aggregation independent of adenosine diphosphate. In Prostaglandins in Hematology (Edited by Silver M. J., Smith J. B. and Kocsis J. J.), pp. 247-276. Spectrum Publications, New York, U.S.A. Pollock W. K., Rink T. J. and Irvine R. F. (1986) Liberation of [rH]arachidonic acid and changes in cytosolic free calcium in fura-2-loaded human platelets stimulated by ionomycin and collagen. Biochem. J. 235, 869-877.