Lysine binding to activated human platelets and its similarity to fibrinogen binding

Btochimica Elsevier

et Biophysicu Actu 847 (1985) 293-300

293

BBA 11592

Lysine binding to activated human platelets and its similarity to fibrinogen binding Galila Agam ‘, Rivka Luria ‘, Orit Shohat ‘, Alexander Uri Seligsohn ’ and Avinoam Livne “.* u Department

of Btolog) and h Department of Hemutologv, Ben Gunon Uniuemty of the Negw, MedrcuI School, Tel Autr~ Uniuersrty, Tel AC (Isruel)

(Revised

Key words:

Lysine binding;

(Received manuscript

Fibrinogen

Dvilansky Beer Shwu.

‘. und

’ Suc~klrr

April 23rd, 1985) received August 26th. 1985)

binding;

Thrombasthenia;

(Human

platelet)

Platelet surface glycoproteins IIb-IIIa are considered to function as the binding site for fibrinogen. Fibrinogen binding is essential for platelet aggregation and several amines have been shown to inhibit this binding. The present study compares the binding properties of 12’I-fibrinogen and 1‘Hllysine with platelets activated by the Ca2+ ionophore A23187. Many lines of similarities in the binding properties are apparent; however, several differences were also found. The similarities are listed below and the differences are pointed out in parentheses. (a) Marked enhancement by platelet activation; (b) deficiency of binding by thrombasthenic platelets lacking the glycoproteins IIb-111% (c) saturability (fibrinogen binding approaches saturation at more than 12 PM, within 10 min; lysine binding at more than 100 mM within 1 min); (d) Ca2+-dependence (at 1 mM Ca2+ lysine binding is minute and fibrinogen binding is half-saturated); (e) reversibility; the binding achieved within 10 min is exchangeable; dissociation depends upon time and external ligand concentration; (f) inhibition by the oligoamines His-Lys and Lys,; (g) inhibition by serum from a thrombasthenic patient who developed anti-glycoproteins IIb-IIIa antibodies; (h) specificity; alanine neither binds to activated platelets nor inhibits fibrinogen binding; it thus appears that the lysine which associates with activated platelets is mostly bound onto the surface of the ceils rather than being incorporated; Moreover, the major site of lysine binding seems to be the complexed glycoproteins IIb-IIIa.

Introduction Fibrinogen binding to activated platelets is a well-established phenomenon, proven to be a prerequisite for platelet aggregation [1,2]. Amino sugars and basic amino acids were shown to inhibit both platelet aggregation [3] and fibrinogen binding to activated platelets [4]. In previous studies we have demonstrated that several series of basic oligoamines (di-, tri-. tetra- and pentalysine, the polyamines putrescine, spermidine and sper* To whom correspondence 0167.4889/85/$03.30

should be addressed

0 1985 Elsevier Science Publishers

mine, etc.) inhibit, in a similar pattern, two platelet functions: (a) platelet aggregation [5.6]; (b) fibrinogen binding to activated platelets [6,7]. Furthermore, an apparent competition between tetralysine and platelet or fibrinogen concentration was shown. As fibrinogen-lysine or fibrinogen-pentalysine association could not be detected, it was assumed that the mode of action of the basic oligoamines involves their interaction with the platelet surface [7]. In the present study attempts have been made to characterize further the interaction of the inhibitory amine with activated platelets. It is shown

B.V. (Biomedical

Division)

that platelet activation leads to enhanced lysine binding. Fibrinogen and lysine binding exhibit remarkable similarity. Evidence is presented that lysine binds to the fibrinogen binding site. Methods

Prepurution of plutelet suspension. Blood was collected into acid citrate dextrose. Platelet-rich plasma was obtained by centrifugation at 120 x g for 10 min. Platelets were separated by gel filtration as follows: first, the concentration of platelets in platelet-rich plasma was increased 3-fold following centrifugation at 200 x g for 10 min so that the platelets did not pack into a pellet; then 1 ml of concentrated platelet-rich plasma was passed through a Sepharose 2B column (Pharmacia Fine Chemicals, Uppsala; 0.8 x 12 cm). The column was equilibrated with Ca’ I- and Mg’ +-free Tyrode solution, pH 7.4, supplemented with 3.6 mg/ml albumin (bovine, from Hyland, Costa Mesa. CA). Platelets were eluted with the same solution. Radioactk~e labeling. Fibrinogen (human, 97% clottable, from Cutter. Berkeley, CA) was iodinated with [“‘lliodine (sodium iodide, “‘1 13-17 mCi/ pg iodide, 100 mCi/ml) with iodogen (Sigma, St. Louis, MO) according to Knight et al. [8]. Separation of the radiolabeled fibrinogen from the unbound “‘1~~ was performed by gel-filtration through a Sephadex G-25 column (Pharmacia; 0.8 x 16 cm). Clottability of the radioactivity in the fibrinogen preparation was 80-84%. Trichloroacetic acid-insoluble material was 95%. Reduced and nonreduced SDS-gel electrophoresis of the “‘l-fibrinogen revealed the typical bands of fibrinogen. indistinguishable from those of the unlabeled preparation. Binding ussuys. The binding of “51-fibrinogen to washed platelets was measured as follows: gelfiltered platelets ((2.5-7.5). 10’) were supplemented with 0.875 mg “51-fibrinogen. and either 2.5 mM CaCl, or 0.2 mM EDTA in a final volume of 0.35 ml, and activated by 10 I_LMADP (Sigma. St. Louis, MO). or 1.25 pg A23187 (Ca’* ionophore. from Calbiochem AG. Luzern). The suspensions were incubated at 23°C. without stirring, for 10 min, and then aliquots (100 ~1, in triplicates) were layered over 1 ml of 20% sucrose in a 1.5 ml conical tube and spun for 1.5 min at

12 000 rpm in a Beckman microfuge. The “‘Ifibrinogen bound was determined by counting the radioactivity in the tip of the centrifuge tube. following the removal of the supernatant and the detachment of the tip. The total “51-fibrinogen association was corrected by subtracting the nonspecific association, i.e. the counts recorded in the absence of Ca’ ’ and A23187, to obtain ‘fibrinogen binding’. The binding of r.-[4,5-3H]lysine (monohydrochloride), or r.-[4-‘4C]alanine (8.5 Ci/mmol, 1 pCi/pl, and 165 mCi/mmol, 50 pCi/ml, respectively, both from the Radiochemical Centre, Amersham, U.K.) to gel-filtered platelets was measured as described for fibrinogen binding, except that fibrinogen was replaced by either lysine or alanine (Sigma, St. Louis, MO). The final concentration of the amino acids was 50 mM, unless otherwise indicated. Where indicated, His-Lya or tetralysine (Lys4) (both from Miles-Yeda, Rehovot) were added to the assay mixture just prior to the platelets. Platelet uggregution. Aggregation of platelet-rich plasma in citrate was followed turbidometrically in a Payton Aggregometer, in a final volume of 0.8 ml. Activation was induced by 10 PM ADP or 1.25 pg A23187 r.-L.vsine dissociation. The procedure was essentially as described by Peerschke [9] for the measurement of fibrinogen dissociation. Gel-filtered platelets (lo”/ ml) were incubated at room temperature with 3.75 pg/ml A23187 and 0.1 M lysine (85 Ci/mmol). After 10 min. samples of 100 ~“1 were (A) centrifuged through 1 ml of 20% sucrose in Tyrode’s solution, to measure total binding; (B) diluted IO-fold with Tyrode containing 3.75 pgg/ml A231 87 and different concentrations of unlabeled lysine. After 1. 5 and 15 min intervals, lOO+l aliquots of the preparations (B) were centrifuged through 1 ml of sucrose solution to measure the amount of remaining bound lysine. The tips containing the pellets were counted in a P-counter. Glunzmunn’s thrombustheniu putients. Criteria for establishing the diagnosis were: a life-long severe bleeding disorder. absent clot retraction, a prolonged bleeding time (> 15 min) and absent ADP-induced platelet aggregation. In addition, less than 5% of normal binding of a monoclonal antibody against glycoproteins IIb-IIIa to platelets

was obtained (Caller B.S.. Peerschke, E.I., Seligsohn. U., Scudder, L.E., Nurden, A.T. and Rosa, J.P., unpublished data). Results L+ine ussociution with platelets Table I compares the association with platelets in the presence or Ca’ ’ ionophore A23187 or ADP. nificantly more lysine associates than with non-activated platelets. between the values obtained with non-activated platelets is henceforth binding’.

of [‘Hllysine absence of the As shown, sigwith activated The difference activated and termed ‘lysine

TIME,

Fig. 1. Time dependence binding to A23187-activated

Comparison of (ysine and fibrinogen binding As shown in Fig. 1, both fibrinogen binding and lysine binding to A23187-activated platelets reach saturation, but at different times: fibrinogen binding is saturated after 5 min of incubation, while that of lysine binding approaches saturation within 1 min. Fig. 2 depicts the data of the binding of [ ‘Hllysine (A) and ‘2SI-fibrinogen (B) to gel-filtered platelets. The data are plotted as recommended by Klotz [II]. for systems in which many receptor sites exist on a single binding entity. A tendency towards saturation is achieved at 100 mM lysine and over 11.8 PM fibrinogen. The similarity between the two curves is noteworthy. K, values estimated from these data are 5 I_IM and 50 mM for fibrinogen and lysine binding, respectively. Both fibrinogen binding and lysine binding to A231 87-activated platelets are totally dependent upon added Ca’+. At 1 mM Ca2+ fibrinogen binding is already half-saturated, while lysine binding

TABLE

[Lysine]

ASSOCIATION

are means+

WITH

and

[ ‘Hllysine

Ln

=

> i

LYSINE

mM

FIBRINOGEN

L

p’M

Fig. 2. Dependence of “51-fibrinogen and [‘Hllysine binding to A23187-activated platelets upon ligand concentration. plotted according to Klotz [ll]. Specific activity in the assay mixtures was 0.08 pCi/mg and 3 pCi/mmol for fibrinogen and lysine. respectively.

is minute. mM Cal+ ence upon by 10 PM

Both (Fig. Ca2+ ADP

processes are saturated above 2.5 3). The same pattern of depend1sapparent with platelets activated (data not depicted).

S.E. n indicates

Activation

by

PLATELETS the number n

(mM)

of experiments.

A23187.3.75 pg/ml A23187. 3.75 pg/ml ADP. 10 PM

2 10 4

each In triplicate

Lysine associated (nmol/lOx cells) non-actwated

10 50 50

of “’ I-fibrinogen platelets.

I

[‘HILYSINE Results

min

87i13 8Oi17 8Oi17

(a)

with platelets

Lysine binding (b ~ a) cells)

P. b vs. a

5Ok 23 218+II3 111* 51

< 0.05 < 0.05 < 0.01

(nmol/lOx

activated 137ilO 298 + 96 191* 34

(b)

TABLE

II

OLIGOAMINES INHIBIT THE BINDING A23187-ACTIVATED PLATELETS

OF LYSINE

TO

[ ‘H]Lysine concentration was 50 mM and the control binding was 222 + 70 nmol/lOx cells. Results are means k SE. Number of experiments (each in triphcate) is indicated in parentheses. * Significantly different from control (P < 0.001) and from each of the other results in the table (P < 0.001). Statistical analyses were performed using Student’s l-test. Oligoamine

HwLya His-Lys Tetralysine

(mM)

Inhibition (% of control)

10 20 5

43+7 (4) 50*5 (3) 85*5(12) *

Concentration

Fig. 3. Effect of added Ca” upon ‘251-fibrmogen and [ ‘Hllysine binding to A23187-activated platelets. Results are means of three experiments k S.E.

Dissociation of bound lubeled (wine from A23187uctiuated plutelets The dissociation of [jH]lysine from A23187treated gel-filtered platelets was evaluated as a function of receptor occupancy in accordance with

mM

100

in

LYSINE

:

medium

0

25 30

the protocol already applied for fibrinogen binding [9] in order to differentiate between lysine uptake and binding. Lysine dissociation was measured after lo-fold dilution with buffer containing 0, 25 or 50 mM unlabeled lysine. As shown in Fig. 4, the dissociation increases with increasing lysine in the diluting buffer, and it is biphasic. With buffer alone 45% of labeled lysine dissociates in 15 min while with 25 and 50 mM lysine 66% and 9.5% of the bound lysine, respectively, dissociate. The results are thus compatible with reversible binding rather than incorporation into the cells.

l

4

Inhibition of platelet-Jvsine interuction 6-v oligoumines As previously described [7], fibrinogen binding is inhibited by lysine, His-Lys and tetralysine with an increasing order of potency, more than accoun-

10

TABLE

COMPARISON BETWEEN LYSINE AND ALANINE SOCIATION WITH ACTIVATED PLATELETS

3

1

III

I _,

5 TIME,

I

I

10

15

min

Fig. 4. Effect of time and unlabeled lysine on the dissociation of [ ‘Hllysine bound to A23187-activated platelets. mM lysine in medium refers to the concentration of lysine in the diluting Tyrode’s solution.

AS-

The binding with A23187-activated platelets after subtraction of the association with non-activated platelets (140+42 and 305 +20 nmol/lO” platelets for lysine and alanine, respectively); Lys, did not affect these values. Number of experiments (each in triplicate) is given in parentheses. Amino acid t.-Lysine L-Alanine

Bindmg (nmol/lOx

Inhibition platelets)

222+70(13) 14* 1 (2)

(%) 83 + 8(12) 0 (2)

by Lys,

297

TABLE

IV

FIBRINOGEN

AND

LYSINE

BINDING

TO A23187-ACTIVATED

THROMBASTHENIC

PLATELETS

Control values were (for fibrinogen and lysine, respectively) non-activated platelets: 1.45 +0.2 pmol/lOx cells and 8Oi 17 nmol/lO’ cells: activated platelets: 4.45 +0.3 pmol/108 cells and 330+ 18 nmol/lO’ cells; binding: 3.OiO.5 pmol/lO’ cells (18OOOk 3000 molecules/pi) and 250 f 35 nmol/lO’ cells (1.5. lo9 f 2.1 .lOs molecules/pi). These values are in accordance with those presented in Tables I and III and Figs. 1 and 3. Fibrinogen (pmol/lOx

Patient

1 2 3 4

associated cells)

with platelets

non-activated

activated

1.44 1.69 2.15 3.27

1.68 1.69 2.15 3.27

Fibrinogen binding (pmol/lOx

Lysine associated (nmol/lO” cells) cells)

0.24 0 0 0

table for by an additive effect of the amino components. Similarly, Table II shows that on a molar equivalent basis, tetralysine inhibits lysine binding much more than His-Lys. [ ‘H]Lysine binding, measured at a lysine concentration of 59 pM, was 0.11 fmol/lOX platelets. Fibrinogen, at 34.5 PM, completely inhibited this binding. “‘I-fibrinogen binding, measured at 0.18 PM fibrinogen, was also abolished by 34.5 PM unlabeled fibrinogen. Fur-

ik

I

I

1

11

IllIll

3 TIMES

Fig.

1

10 OF

SERUM

1

llllh.?

30

100

DILUTION

5. Similar inhibition of binding of ‘251-fibrinogen and to A2318J-activated platelets by thrombasthenic serum containing anti-platelet activity: effect of serum dilution. 2 ml of serum were preincubated with 5 units of hirudin (Sigma, St. Louis, MO) for 10 min. Then the serum was diluted in the Tyrode’s buffer used for the binding assays. It was added to the assay mixtures just prior to the platelets. Each experimental point reflects the comparison with the binding obtained in the presence of control serum which was treated and diluted in the same way as the thrombasthenic serum. Control serum did not affect the binding. Results are means of four experiments + S.E.

[ ‘Hllysine

with platelets

non-activated

activated

56 6 21 61

91 31 21 73

Lysine binding (nmol/lOx

cells)

35 25 0 12

thermore, fibrinogen at 24.2 PM completely inhibited [3H]lysine binding measured at 0.1 and 1.0 mM (control values 2.2 and 6 nmol/108 platelets. respectively). In an earlier paper [7] we have shown that the inhibition of fibrinogen binding by Lys, was gradually overcome by increasing concentrations of fibrinogen. Lineweaver-Burk plots of these data indicate an apparent competitive inhibition. Comparison of lysine and alanine binding The binding of two amino acids, lysine and alanine, to A23187-activated platelets was compared in order to evaluate the specificity of lysine binding. As clearly demonstrated in Table III, alanine association with activated platelets extends to less than 10% of that of lysine. This finding also indicates that the increase in lysine binding due to activation does not result from increased ‘trapment space’. Furthermore, the association of alanine with activated platelets is not inhibited by 5 mM tetralysine. Fibrinogen and lysine binding to thrombasthenic platelets Platelets derived from four thrombasthenic patients and from four healthy subjects were assayed for aggregation, fibrinogen binding and lysine binding. Platelets were stimulated by A23187. Fibrinogen binding to patients’ platelets ranged between the O-8% of controls, and those of lysine binding, O-14% (Table IV). Similarly, platelet aggregation was also negligible.

29x

Inhibition by Serum As a result of multiple transfusions. one thrombasthenic patient (No. 2) developed a serum activity which inhibits the interaction between platelets and fibrinogen-coated beads. Serum derived from this patient totally inhibited both fibrinogen and lysine binding to A231 87-activated platelets from healthy volunteers. Essentially identical dose-dependence was apparent for the inhibition of lysine binding and fibrinogen binding (Fig. 5). Discussion

The present study demonstrates for the first time that activation of platelets by ADP or A23187 leads to a marked increase in the extent of association of lysine with platelets. This phenomenon adds to the many platelet membrane events which are already known to be affected by activation [12- 231. Association of lysine with non-activated platelets may result from several causes. The outer surface of platelet plasma membranes carries a net negative charge, contributed by sialic acid residues, phosphoryl groups of phospholipids and carboxyl groups of acidic amino-acid residues of proteins [24,25]. The positively charged lysine may interact electrostatically with these groups. Furthermore, procedural aspects, such as some spontaneous platelet activation, trapping of [ 3H]lysine in the platelet sediment, etc.. may contribute to some of the label associated with the non-activated platelets (‘blank values’, amounting to 40-50s of total). Therefore, the amount of lysine detected in non-activated platelets was subtracted to calculate the net association with activated platelets, namely, ’ lysine binding’. Binding to the surface and uptake into platelets are alternative explanations for the increased association of lysine with activated pletelets. Several lines of evidence indicate that binding to the surface is the dominant factor. (a) Maximal specific lysine association is reached after 1 min of incubation. Equilibrium of uptake processes into blood cells is usually achieved after much longer periods of time [26-281. (b) The association of lysine with activated platelets was inhibited by the dipeptide His-Lys: 40% of inhibition of lysine binding was

observed at 10 mM His-Lys, one-fifth the concentration of lysine. Furthermore, tetralysine inhibited the binding much more effectively than the dipeptide. The extended inhibition by tetralysine cannot be explained merely by an additive charge effect, but, apparently, it reflects higher affinity for the site of lysine binding. (c) While I>-lysine binds to activated platelets, the binding of another amino acid, t-alanine, is minute (15-fold less). (d) Labeled t.-lysine that was bound to activated platelets readily dissociated when the platelets were incubated in a diluting buffer containing unlabeled lysine. The percentage of dissociation increased with increasing external concentration of lysine in the buffer and amounted to 95% at 50 mM lysine. (e) Activated thrombasthenic platelets were defective in lysine binding. The present study deals with platelets activated by the Ca” ionophore A23187. The K, value obtained for fibrinogen binding in this system is, approximately, 5 PM. The values reported for platelets activated by ADP are either similar [29,30] or lower [1,31]. Lysine binding resembles fibrinogen binding to activated platelets by several major criteria: kinetic properties. sensitivity to inhibitors and the behaviour of genetically disordered platelets. Kinetic properties The two binding processes show similarities in terms of the dependence upon time, ligand concentration and Ca” added. Binding of fibrinogen and lysine is saturable within minutes, thereafter the ligands dissociate readily. Klotz plots [ll] of the two binding processes as a function of the logarithm of ligand concentration have a similar shape (Fig. 2). Moreover, there is an absolute dependence of the two binding processes upon added Ca’ +. Several intriguing differences in the kinetic properties should be considered. Under the conditions assayed, lysine binding is faster. This may be explained by the huge difference in the molecular size of lysine and fibrinogen. Furthermore, fibrinogen binding may require a conformational change of the polypeptide to fit its receptor. Fibrinogen exhibits a much higher affinity for the platelets. It may indicate the existence of multiple interactions of fibrinogen with the complexed gly-

299

coproteins IIb-IIIa. One of these assumed interactions could involve the positively charged amino acids in the segment of fibrinogen, as demonstrated by Hawiger et al. [32,33]. It is possible that the interaction with lysine reflects this particular interaction. Lysine binding is over four orders of magnitude higher (in moles/platelet) than that of fibrinogen. Since, based on the evidence presented in this study, the major site for lysine binding is the fibrinogen binding site, this discrepancy may be partly explained by the likelihood that this site binds numerous lysine molecules. However, a full explanation is still lacking. The difference in the dependence on Ca2 + concentration may be related to the differences in quantity and affinity between lysine and fibrinogen binding. Sensitivity to inhibitors Lysine binding is inhibited by fibrinogen and fibrinogen binding is inhibited by lysine. The binding of fibrinogen and lysine is similarly inhibited by lysine oligomers. As has already been suggested [5-71, the difference in potencies of the oligopeptides indicates cooperative interactions of the positively charged groups with the platelet membrane. Genetically disordered platelets Thrombasthenic platelets are defective in fibrinogen binding and aggregation [34]. The present communication expands these findings to A23187-activated platelets. The molecular basis of this genetic disorder is a deficiency [35] or an abnormality [36,37] in glycoproteins IIb-IIIa. This molecular disorder relates concomitantly to the binding of fibrinogen and lysine: thrombasthenic platelets, which failed to bind fibrinogen, also exhibited poor lysine binding. Furthermore, serum from a thrombasthenic patient, containing an anti-platelet activity, inhibited identically both binding processes. On this basis, and the supporting evidence discussed above, it is concluded that the fibrinogen receptor is the major site of lysine binding on the outer surface of activated platelets. To sum up, an enhanced association of lysine to activated platelets is demonstrated. This association is mostly due to surface binding, and ex-

hibits marked similarity to fibrinogen binding. Evidence is presented that lysine binds to the fibrinogen binding site. Further study of the binding of defined peptides by activated platelets, as approached in the present work, may serve towards characterization of the detailed structure of the complementary segments of fibrinogen and its platelet receptor. References

1 Marguerie, G.A.. Plow, E.F. and Edgington. T.S. (1979) J. Biol. Chem. 254. 5357-5363 2 Peerschke, E.I., Zucker, M.B., Grant. R.A.. Egan, J.J. and Johnson, M.M. (1980) Blood 55. 841-847 3 Gartner. T.K.. Williams, D.C., Minion, F.C. and Phillips. D.R. (1978) Science 200. 1281-12X3 4 Kinlough-Rathbone, R.L.. Chahil. A.. Perry. D.W., Packham. M.A. and Mustard. J.F. (1979) Blood 54 (suppl.), 249a 5 Agam, G.. Gartner. T.K. and Livne. A. (1984) Thromb. Res. 33. 245-257 6 Livne. A.. Gartner. T.K. and Agam, G. (1983) Thrombos. Haemostas. 50. 132 7 Agam, G., Luria. R., Shoshani, L. and Livne, A. (1985) in Fibrinogen. Structural Variants and Interactions (Henschen. A.. Hessel, B.. McDonagh, J. and Saldeen. T.. eds.), pp. 379- 386. Walter De Gruyter & Co.. Berlin 8 Knight, L.C.. Budzynski. A.Z. and Olexa, S.A. (1981) Thromb. Haemostas. 46, 593-596 9 Peerschke, E.I.B. (1982) Blood 60. 973-978 10 Caller, B.S.. Peerschke. E.I.. Scudder. L.E. and Sullivan, C.A. (1983) J. Clin. Invest. 72. 325S338 11 Klotz. I.M. (1982) Science 217. 1247-1249 12 Hardisty, R.M. and Hutton. R.A. (1966) Br. J. Haematol. 12. 764-776 13 Rittenhouse-Simons. S. and Deykin. D. (1977) J. Clin. Invest. 60. 495-498 14 Deykin. D. (1976) in Platelets in Biology and Pathology (Gordon. J.L.. ed.), Vol. 1, pp. 111-119. North-Holland Publishing Company. Amsterdam 15 Smith, J.B. and Silver, M.J. (1976) in Platelets in Biology and Pathology (Gordon, J.L., ed.). Vol. 1. pp. 331-352, North-Holland Publishing Company, Amsterdam 16 Kaulen, H.D. and Gross, R. (1976) Thromb. Haemostas. 35, 364-376 17 Nathan, I.. Fleischer, G., Livne, A.. Dvilansky. A. and Parola, A.H. (1979) J. Biol. Chem. 254, 9822-9828 18 Nachman. R.L. and Leung. L.L.K. (1982) J. Clin. Invest. 69, 263-269 19 Bennett, J.S., Vilaire. G. and Clines. D.B. (1982) J. Biol. Chem. 257. 8049-8054 20 Gartner, T.K., Gerrard, J.M.. White. J.G. and Williams. D.C. (1981) Blood 58. 153-157 21 Jaffe, E.A., Leung, L.L.K.. Nachman, R.L.. Levin. R.I. and Mosher. D.F. (1982) Nature 295, 246-248 22 Gartner, T.K.. Doyle, M.J. and Mosher. D.F. (1983) Thromb. Haemostas. 50, 124

300

23 Rotman, A., Makov, N. and Lischer, E.F. (1982) Proc. Nat]. Acad. Sci. USA 79, 4357-4361 24 Stoltz. J.F. and Nicolas, A. (1979) Blut 38, 103-117 25 Tanoue, K., Jung, SM.. Isohisa, I., Yamamoto. N. and Yamazaki, H. (1982) Acta Haematol. Jpn. 45. 1391-1400 26 Agam, G., Gasner, S.. Bessler. H.. Fishman, P. and Djaldetti, M. (1976) Br. J. Haematol. 33, 53-59 27 Bessler, H., Agam. G. and Djaldetti, M. (1976) Thromb. Haemostas. 35, 350-357 28 Malik, Z., Agam, G. and Djaldetti, M. (1979) Acta Haematol. 61, 138-143 29 Kornecki, E., Niewiarowski, S.. Morinelli. T.A. and Kloczewiak, M. (1981) J. Biol. Chem. 256, 5696-5701 30 Marguerie, G.A., Thomas-Maison, N., Larrieu. M.J. and Plow, E.F. (1982) Blood 59, 91-95

31 Bennett, J.S. and Vilaire. G. (1979) J. Clin. Invest. 64. 1393-1401 32 Hawiger, J.. Kloczewiak, M., Timmons. S.. Strong. D. and Doolittle, R.F. (1983) Ann. N.Y. Acad. Sci. 408, 521-535 33 Kloczewiak. M.. Timmons, S., Lucas. T.J. and Hawiger, J. (1983) Clin. Res. 31, 534a 34 Mustard, J.F.. Kinlough-Rathbone, R.L., Packham, M.A., Perry. D.W., Harfenist. E.J. and Pai. K.R.M. (1979) Blood 54, 987-993 I 35 Nurden, A.T. and Caen. J.P. (1974) Br. J. Haematol. 2X. 253-260 36 Lightsey, A.L., Thomas, W.J.. Plow, E.F.. MacMillan, R. and Ginsberg, M. (1981) Blood 58 (Suppl. l), 199a 37 Nurden, A.T.. Rosa, J.P., Boizard, B.. Didry. D., Legrand, C. and Parquet, C. (1983) Thromb. Haemostas. 50, 216