Shear stress activation of platelets with subsequent refractoriness

THROMBOSIS RESEARCH 45; 29-37, 1987 $3.00 + .OO Printed in the USA. 0049-3848/87 Copyright (c) 1987 Pergamon Journals Ltd. All rights reserved.

SHEAR STRESS ACTIVATION OF PLATELETS WITH SUBSEQUENT REFRACTORINESS

Ivar Aursnes, Jon Sundal and Terje Nome Department of Medicine, Aker Hospital, Oslo , Norway (Received 1.5.1986; Accepted in revised form 9.9.1986 by Editor F. Brosstad)

ABSTRACT Platelet rich plasma was prepared with soybean trypsin inhibitor (SBTI) as anticoagulant, to preserve normal and subsequently tested in an Ca++-concentration, aggregometer. Shear stress alone, induced by stirring, resulted in platelet aggregation and disaggregation. Such platelets showed only minor aggregational response on the addition of ADP, indicating that the platelets had been partly desensitized towards ADPstimulation. When the stirrer was started at the same time as the addition of 1.6 pM ADP, the aggregation velocity was related to the speed of the stirrer. Disaggregation always started 1.5 min after the stimulus. Platelet shape change was apparently not influenced by the stirring. Since platelets from SBTIanticoagulated blood could be aggregated by stirring of alone, we examined the effect of centrifugation and 10 min. Signs of whole blood at 50xg for 3 transient aggregation and release of alpha-granule content were respectively observed. INTRODUCTION reaction is avoided, the typProvided the platelet release ical response of platelets in platelet rich plasma in an aggregation followed, during the aggregometer is a quick (1). This aggregation ensuing minutes, by a disaggregation induces shear stress and collisions requires stirring, which between platelets, and also requires a chemical reagent like adenosine diphosphate (ADP). It is not known whether shear is the major contributing mechanical factor, as stress _---_------------

Key words: Adenosine diphosphate platelet shape change, refractory

(ADP), platelet platelets 29

aggregation,

30

ACTIVATION AND REFRACTORINESS

Vol. 45, No. 1

collisions between platelets could alone be responsible for the effect of stirring. The experiments to be reported were performed with the following deviations from the commonly used procedures. The blood was anticoagulated without lowering extracellular Ca++concentration. This may render the platelets more sensitive to external stimuli; probably more similar to the in vivo situation. It is known that platelet rich plasma can hardly be prepared without activation of the platelets during the proce(2). We dure, even with the use of citrate as anticoagulant therefore minimized the platelet activation through ADPscavenging and prostaglandin synthesis inhibition. ADP, when to a fixed final concentration. We added, was standardized found that the degree of platelet responses were inversely related to the pre-treatment of the platelets as regards induction of shear stress. MATERIALS AND METHODS Blood sampling. Human blood was used throughout the study. The blood was drained from an antecubitalvein either without stasis, or using minimal stasis. The first two ml were discarded and thereafter the blood was sampled through a wide bore needle (gauge 1.2 mm) directly into polypropylene tubes prefilled with one tenth of the final volume with saline either containing the anticoagulant or sometimes platelet antiactivating drugs as well. The tubes were capped and inverted three times. Anticoagulation. Trypsin inhibitor from soy beans (SBTI) ourchased from Siqma Chemical Comoanv, USA, was used for anticoagulation of the blood. A final-concentration of 3 mg/ml was found to prevent formation of fibrinopeptide A when tested by radio-immunoassay in plasma samples at the end of an experiment. Alternatively 0.11 mol/l citrate to l/10 of blood Platelet-rich plasma volume was used for anticoagulation. (PRP) was prepared from anticoagulated whole blood by centrifugation at 270xg for 15 min at room temperature. Inhibition of prostaglandin synthesis was achieved by --adding a small quantity of 10 % acetyl-salicylic acid (ASA) dissolved in ethanol to a final concentration of 100 umol/l ASA and 0.01% ethanol in saline. Aqqreqometer tests The Payton aggregometer (Ontario, --Canada) was used for testing the effects of varying the speed of the stirrer, and the aggregometer tests with constant stirrer speed were performed with the Born/Michal aggregometer (London, England). Incubation in the aggrego eter for 3 min prior to the test secured a temperature of 37d" C at the start of the experiment. Contrary to common practice, stirring was not performed during the incubation period. The stirrer was started together with the addition of ADP at a final concentration of 1.6 uM in the platelet-rich plasma. In some series of experiments an ADP scavenging system was added

Vol. 45, No. 1

ACTIVATION AND REFRACTORINESS

The scavenger consisted of together with the anticoagulant. phosphoenol pyruvate (PEP, 20 pg/ml blood) and pyruvate kinase (PK, 10 U/ml blood, both from Sigma Chemical Company, USA), which tended to transfer free ADP in plasma into ATP. Platelet countinq. This was performed with a whole blood cell counter from Analys Instrument, Stockholm, Sweden. which also measured HCT. Therefrom the platelet concentration in plasma was calculated. Platelet stimulation by centrifugation. Ten ml of anticoagulated bsrn?a 12.5 ml, 11 mm wide polypropylene tube was put into a swing-out centrifuge immediately after withdrawal, and spun at room temperature and 5Oxg(,.+) for 3 min. Braking was not used. Platelet counts were done in plasma samples more or less admixed with red cells, which were removed by a 20 ul disposable glass capillary pipette successively from positions 20, 40 and 60 mm from the top of the tube and mixed with 4 mlof 23 mM EDTA in buffered saline. In other tests the centrifugation time was increased in order to obtain a detectable release of material located in the alpha-granules (3). After a 10 min spin at 5oxq(m x)' 2 ml of the platelet-rich plasma from the top layer was 0 ested for supernatant content of beta-thromboglobulin and compared with plasma samples taken from the blood before the centrifuge procedure. The samples were immediately added to 0.5 mlof a pre-coole$ EDTA/prostaglandin El solution (4), and centrifuged at 4 C and 1200xg for 20 min. The mid-portion of the platelet poor plasma was centrifuged for another 20 min before assay of content of beta-thromboglobulin (RIA from Amersham, England), a specific marker of alpha-granule release from platelets. RESULTS The effect of stirring platelet rich,plasma in the aggregometer. The platelets aggregated when stirring was started. Typical examples chosen from 3 similar experiments are shown in Fig. 1. No aggregation could be observed if PRP was prepared with c'trate as anticoagulant, or if the samples were preheated to 37 6C in the aggregometer 3 min prior to stirring. Moreover, the length of unstirred pre-incubation time of PRP at room temperature influenced the outcome, with the most brisk responses in the middle samples, when three tests were performed in succession during one half hour. Fig. 1 demonstrates that aggregation could be induced without ADP, and also that subsequent addition of ADP produced only a small response. A positive control experiment is included in the bottom tracing. Such a response could be obtained when stirring was started together with addition of ADP, irrespective o the time lag before testing and of length of incubation at 375 C.

31

ACTIVATION AND REFRACTORINESS

32

stirring

Vol. 45, No. 1

ADP

ADPa

mrring

FIG. 1 with obtained 1 ml of SBTI-antiAggregometer tracings coagulated platelet-rich plasma after 8, 17 and 26 min. of pre-incubation at room temperature. Magnetic stirring was initiated simultaneously with the transfer of the plasma sample to the aggregometer. ADP (2Oul) at a final as indicated. The lowest cont. of 1.6 pM was added tracing is from a control experiment after 3 min. preincubation at 37' C. ADP was added simultaneously with the start of stirring. Effect of varyinq the speed of stirring on aqgregometer responses. A typical experiment from a series of five is shown in the top part of Fig. 2. These experiments were performed in platelet rich plasma prepared with the intention to obtain maximally sensitive platelets, i.e. ADP released during the preparation was scavenged by PEP/PK. (Such platelets did, however, not aggregate spontaneously when tested as in Fig. lb. From Fig. 2 it can be seen that aggregatory responses are

Vol. 45, No. 1

ACTIVATION AND REFRACTORINESS

greatly influenced by the speed of the stirrer. The experiments also show that shape change, represented by the initial upward deflection of the trace, was probably not influenced by the stirring, since an equal response was obtained with and without stirring at 600 r.p.m.

V

Max.

FIG. 2 Aggregometer tracings obtained by stirring 0.7 ml of platelet-rich plasma with a metal rod at 0, 600 and 1200 r.p.m., and at maximum speed. Stirring was started simultaneously with the addition of 20 gl ADP to a final concentration of 1.6 uM. PRP was produced either from blood anticoagulated with SBTI or citrate as indicated, and PEP/PK as ADP- scavenger was always added to prevent ADP refractoriness, and acetyl-salicylic acid to inhibit prostaglandin formation.

33

34

ACTIVATION AND REFRACTORINESS

Vol. 45, No. 1

The disaggregation phase always started about 1.5 min after the start of the stirring, even when the ADP was added several minutes before the stirrer was activated. The bottom part of Fig. 2 shows similar results as above, with citrate as anticoagulant. The aggregation responses, however, were diminished compared with SBTI-anticoagulated PRP.

t SAMPLE

1

2 EDTA

1

2 SBTI

1

2 ADP

FIG. 3 Concentrations of platelets in whole blood anticoagulated with EDTA (control experiment), SBTI, or SBTI with ADP (positive control experiment) after centrifugation for 3 minutes at 50xg. Sample 1 was taken 40 mm and sample 2 60 mm below the top of the tube and expressed in per cent of the platelet concentration 20 mm from the top.

Platelet activation induced by centrifuqation alone. The results reported above prompted us to examine whether, the effect of centrifugation alone could elicit platelet activation and aggregation in whole blood with normal Ca++cont.. Support for this idea was sought by determining the distribution of platelet concentrations in the centrifuge tube after centrifugation (Fig. 3). Whole blood with normal Ca++cont. (central part of Fig. 3) was compared with blood with inhibited aggregation (left) and with ADP added (right). There are increasing differences in platelet-counts between sample 1

Vol.

45,

No.

1

ACTIVATION

AND REFRACTORINESS

After

Before -

Control experiment

n-5

35

Before After Centrifugation n=9

FIG. 4 Beta-thromboglobulin in plasma samples before and after centrifugation of SBTI- anticoagulated , freshly drawn blood at 50xg for 10 min at room temperature, and in control plasma samples taken from blood which had not been centrifuged. (obtained 40 mm fromthetopofthetube) and sample 2 (60 mmm from the top) when the left, middle and right part of the figure are compared (p=O.O3 for differences between left and middle experiments, Wilcoxon's two-sample test). When the experiment in Fig. 3 (middle part) were performed with 10 min centrifugation time, beta-thromboglobulin was found to be released during the procedure (Fig. 4). DISCUSSION Increased sensitivity towards aggregatory stimuli results when platelets are studied in vitro in surroundings containing normal Ca++ -concentration. Such reactive platelets aggregate when they are stirred and then quickly disaggregate (Fig. 1). Only to a minor extent can repeated aggregation of platelets be elicited when platelets are challenged with ADP some minutes later. Then they are desensitized or refractory (5) to the second stimulus. Apparently this ADP-refractoriness was

36

ACTIVATION AND REFRACTORINESS

Vol. 45, No. 1

induced by the stirring. It is possible, therefore, that the disaggregation phase following aggregation also has been induced by the stirring, i.e. due to refractoriness induced by the stirring. The phase of disaggregation always started 1.5 min after initiation of the reaction (Fig. 2). The figure also indicates that shape change is not affected by the stirring, whereas aggregation is affected. From the present experiments it cannot for sure be decided whether stirring is important by means of providing shear stress collisions between platelets or also by inducing on the platelet membrane leading to further reactions. In support of the second alternative one can argue that, as refractoriness seems to be induced by the stirring , it can possibly also induce other reactions in the platelets, such as activation. The observed activation induced by centrifugation alone, is in support of this alternative. ADP is important for platelet aggregation under the presently described conditions. Addition of an ADP scavenging system prevented spontaneous aggregation, as did incubation of PRP at 37' C before stirring. The latter probably induced breakdown of ADP which had been released during the preparation of PRP. ADP under such conditions may originate from red cells (4). The idea that trace amounts of ADP are necessary for aggregation induced by stirring, is in accordance with previous observations of "preactivation" and "deactivation" of platelets taking place during centrifugation (6). These phenomena were also found to be dependent on trace levels of ADP. The observations as depicted in Fig. 3 are interpreted as follows. The experiments on the left part of the figure give the reference values for platelet concentrations in various layers of a centrifuge tube after centrifugation of platelets which do not aggregate because ionized calcium has been removed. In the middle part of the figure, with normal Ca++concentration, a significantly different distribution of platelet counts is observed: the concentration differences between sample 1 and sample 2 are greater than in the experiments shown on the on the left part of the figure. It is that the platelets have been and likely aggregated disaggregated during the centrifugation, and that this has had an effect on the gravitational constants for the moving particles. As a positive control the same experiments were reproduced with deliberate induction of platelet aggregation by the addition of ADP at the start of the centrifugation. The outcome is shown on the right part of Fig. 3. The difference in platelet counts between sample 1 and sample 2 is then greatly enhanced. The suggestion that the platelets have been activated, and also reversibly aggregated, in the experiments with SBTI without ADP ( middle part of Fig. 3), is to a large extent supported by the findings in Fig. 4. A definite release of alpha-granule content from platelets is observed after stimulation only by centrifugation of SBTIanticoagulated blood.

Vol. 45, No. 1

ACTIVATION AND REFRACTORINESS

37

Shear-induced platelet deactivation has been described in canine platelets (7). By the present report the observed phenomenon is extended to human platelets and to more physiological test conditions, i.e. normal Ca++concentration, lower ADP test-concentration and lower levels of shear stress enforced upon the platelets. Our findings support the view that platelets can be deactivated by mechanical stimulation in vivo (8).

ACKNOWLEDGEMENT A part of this work was performed at the Endocrinology Laboratory, Rikshospitalet, Oslo, Norway, with the skillful assistance of Mrs. Jerild Theie Vinje. The work was supported by The National Council on Cardiovascular Diseases and The Norwegian Diabetic Association. Correspondence to: I.Aursnes M.D., Department UllevAl Hospital, 0407 Oslo 4, Norway.

of Medicine,

REFERENCES 1.

BORN, G.V.R. and CROSS, M.J. The aggregation platelets. J Physiol (Lend) 168, 178-195, 1963.

2.

FISKERSTRAND, C., BURNETT, E., BHATTACHYRYA, S, and Platelet aggregation - A scanning ANDERTON, J.L. electron microscopy study. Haemostasis l2, 54, 1982.

3.

HOLMSEN, H., DAY, H.J. and STORMORKEN, H. The blood platelet release reaction. Stand J Haematol Suppl. & l-26, 1969.

4.

AURSNES, I., GJESDAL K. and ABILDGAARD, U. Platelet aggreinduced by ADP from unsheared erythrocytes at gation physiological Ca++-concentrations. Br J Haematol 47, 149-151, 1981.

5.

HOLME, S. and HOLMSEN, H. ADP-induced refractory state of on aggreplatelets in vitro. I. Methodological studies gation in.platelet rich plasma. Stand J Haematol l5, 96103, 1975.

6.

interaction AURSNES, I. and VIKHOLM V. On a possible between ADP and mechanical stimulation in platelet activation. Thromb Haemost 5l, 54-56, 1984.

7.

RAMSTACK, J.M., ZUCKERMAN, L. and MOCKROS, L.F. Shearinduced activation of platelets. J Biomechanics l2, 113125, 1979.

8.

DALE, J. Reduced platelet adhesiveness in patients with prosthetic ball valves: Relation of adenosine diphosphate and mechanical trauma. Am Heart J 94, 562-657, 1977.

of

blood