THROMBOSIS RESEARCH 60; 33-42,199O 0049-3848/90 $3.00 + .OO Printed in the USA. Copyright (c) 1990 Pergamon Press pk. All rights reserved.
LOCALIZATIONOF THE STRUCTURALDOMAIN RESPONSIBLEFOR THE CHEMOTACTICPROPERTIES OF THROMBINON POLYMORPHONUCLEAR LEUKOCYTES
A. MORIN, M. MARCHAND ARVIER, F. DOUTREMEPUICH", C. VIGNERON Centre de Transfusion sanguine, Avenue de Bourgogne 54500 Vandoeuvre, France +:-Faculte de Pharmacie, 3 Place de la Victoire 33076 Bordeaux Ckdex, France (Received
2.4.1990;
accepted
(Received
in original form 14.5.1990
by Executive
by Editor O.N. Ulutin)
Editorial Office 3.8.1990)
ABSTRACT Human athrombin at 1.1.1O-5 M is chemotactic for human polymorphonuclear leukoc tes. This thrombin pro erty disappears when the athromr M) hirudin (l.32.lO-. P M) mixture is realized. The same bin (l.l.lO-‘ result is obtained when the thrombin at 1.1. lo-" M is inhibited by antithrombin III in a ratio of 1 mol of thrombin for 4.5 mol of antithrombin III. The hirudin and the antithrombin III appear therefore to mask, by their binding the structural domain responsible for the chemotactic properties of thrombin on polymorphonuclear leukocytes.
INTRODUCTION
Thrombin plays a role of prime importance at the level of the hemostatic system. Its action also extends to plasmatic factors of this system as well as to the vascular and even cellular level (l,2,3). However, outside its action on the hemostatic system tithrombin plays a role in non enzymatic reactions (4,5) in particular it has a chemoattractive power on monocytes and certain macrophage like cell lines (6,7,8). This thrombin property is not due to the catalytic site (6,8,9) nor to the fibrinogen recognition site (7,9) of this enzyme. On the other hand, the thrombin, inhibited by antithrombin III (physiological inhibitor of the enzyme) or by hirudin (specific inhibitor of thrombin) no longer acts like a cytotaxin on monocytes (6,7,9). The inhibitors must therefore mask the site or sites responsible for the chemotaxis carried out by the thrombin on monocytes. It has been shown that oithrombin also exerts a chemotactic activity for the polymorphonuclear leukocytes (10,11,12) but the structural domain of the cxthrombin responsible has not yet been distinguished. The aim of this work is to localize it. To this end the
Key words: Chemotaxis, polymorphonuclear leukocytes, thrombin, antithrombin III, hirudin. 33
34
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
chemotactic activity of the octhrombin will be studied in the presence of inhibitors such as hirudin and antithrombin III, whose interaction sites are known.
MATERIALS ANDMETHOD
*Protocol There are two series of experiments in this work: - Study of the chemotactic activity of human octhrombin-hirudin mixture on polymorphonuclear leukocytes. - Study of the chemotactic activity of human octhrombin-antithrombin III mixture on polymorphonuclear leukocytes. The chemotaxis technique under agarose gel was used after isolation of human leukocytes by accelerated sedimentation on dextran. The experiments were carried out under sterile conditions in avoid bacterial order to contaminations. The chemotactic properties of the compounds belonging to the same experimental series were determined in parallel and with regard to the leukocytes of 5 healthy voluntary subjects. *Reagents Dextran T500 (Pharmacia, Sweden) at 2% in solution NaCl at 9%., NH4Cl at 0.87% in distillated water, glucosed phosphate salin buffer (glucdsed PBS)(137 mM NaCl, 3 mM KCl, 8 mM Na2HC03, 1.4 mM KH2P04, 10 mM glucose, pH = 7.41, agarose (indubiose A37, IBF, France), minimum essential medium (MEM)(concentrated 10 times, Gibco, Scotland), human serum inactivated by heat and conserved at -8ooc, NaHC03 at 7.5%, methanol (Cooperation Pharmaceutique Francaise, France), formaldehyde (Prolabo, France), basic fuschine solution at 1% in ethanol, N formyl-L methionyl-L leucyl-L phenylalanine (FMLP)(Sigma, France) at 10-7-M, human tithrombin (Sigma, France), recombinant hirudin (E.Coli), human antithrombin III (CRTS Nancy, France). *Leukocytes isolation Human blood was collected by veinipuncture in heparin tubes (12 VI/ml of blood). This blood was mixed volume to volume with dextran. After sedimentation the supernatant was collected and then 1 volume of supernatant was mixed with 4 volumes of cold NH4C1(40C) for ten minutes in order to haemolyse the red blood cells. This mixture was centrifuged 8 minutes at 230 g and at 20°C. The leukocyte pellet was washed in glucosed PBS then put into suspension in the same buffer. The number of leukocytes was adjusted at 2.107 polymorphonuclear leukocytes/ml. %hkmot&is under agarose gel The technique used was similar to Cutler's (13) and has already been described by Marchand-Arvier (14). - Gel preparation A gel made from: 1.2% agarose, 79% distillated water, 10% MEM, 10% human serum, 1% NaHC03 was deposited on Petri dishes at a height of 5 mm. - The test In the gel some wells a and b (3 mm diameter, 3 mm distance) were cut. For each chemotactic dish 4 pairs of wells a-b are arranged: 1 pair of wells serves as a negative control, the three others for study of the chemotactic properties of the tested compound. The a wells were filled with 10 pl of the leukocyte suspension whereas b wells were filled with 10 pl of the compound to be tested. The dishes were incubated 3 hours at 37OC under 5% of CO2 in a humid atmosphere. The reaction was stopped by adding 5 ml of methanol for 30 minutes and then 5 ml of formaldehyde (30 minutes equally). The agarose gels were unstuck from the dishes and the leukocytes were visualized using fuschine coloration. Chemotactic mobility (CM) and spontaneous mobility (SM) were
Vol. 60, No. 1
COAGULATION ANDPMNCHEMOTAXIS
35
determined under an optical microscope with a calibrated eye piece. From these results a chemotactic index was calculated : Chemotactic index CI =:!$ - The controls They were carried out with the leukocyte suspension used for the test. . Positive control: performed in 3 pair of wells a-b. Wells b were filled with 10 pl of FMLP whilst wells a were filled with 10 pl of leukocyte suspension. . Negative control: made for each chemotactic dish in a pair of wells a-b. Wells a received 10 ~1 of leukocyte suspension whilst wells b remained empty. The negative control represents the spontaneous mobility of leukocytes. To validate the test the chemotactic mobility (CI) obtained for the positive control must be higher than 1.20 and for the negative control they must be inferior to 1.00. 'ATestedsubstances - CIthrombin and hirudin Human cxthrombin preparation has a biological activity of 3313 NIH/mg. Isoelectrofocalisation of this preparation, thanks to the Phast System (Pharmacia, Sweden) with a gradient of pH from 3 to 9 and after coomassie blue staining only shows a single band. In this study, the final thrombin concentration will be of l.l.lO-5 M as a previous study has shown a chemotactic action of thrombin between (?.28.10e5M and 2.2.10e5 M (12). The thrombin-hirudin mixture is obtained by putting into contact a volume of thrombin at 2.2.10 -5 M in PBS buffer with a volume of hirudin at 2.64.10~" M in the same buffer. The inhibition of thrombin was confirmed: 0.5 ml of EDTA plasma (1.8 mg/ml of blood) added with 10 ~1 of thrombin at 1.1.10m5 M coagulated in 15 seconds whilst 0.5 ml of the same plasma added with 10 ~1 of the thrombin-hirudin mixture had a coagulation time superior to 120 seconds. - ccthrombin and antithrombin III A solution of human athrombin at 1.1.10m5 M was carried out in MEM solution. The preparation of antithrombin III used has a biological activity of 0.94 U/'mg.An isoelectrofocalisation on a pH gradient of 3 to 9 on Phast System of the antithrombin preparation was made. The coomassie blue staining reveals a major band corresponding to antithrombin III and some traces of albumin. The thrombin-antithrombin III complex was obtained by putting into contact 1 mol of thrombin with 4.5 mol of antithrombin III for 30 minutes at 37oC (15). The thrombin inhibition was checked, the coagulation time of 0.5 ml of EDTA plasma added with 10 ~1 of thrombin at 1.1.1O-F'M is 15 seconds whereas the coagulation time of 0.5 ml of the same plasma mixed to 10 f11 of the thrombin-antithrombin III mixture was superior to 12@ seconds. %Statistical analysis For each substance tested, 15 CI values were obtained. The different compounds from the same experimental series were compared with each other using the Student t test. p inferior to 0.05 was considered as significant.
RESULTS
*Study of the chemotactic activity of human o~thrombin-hirudin mixture on polymorphonuclear leukocytes (figure 1) For this exoerimental series the oositive control gives a mean CI of 1.88 and a standard deviation of 0.27 whereas the mean CI value of the negative control (or spontaneous mobility) is 0.88 with a standard deviation of 0.08. So the other results obtained in this series could be validated. The human octhrombin
COAGULATION AND PMN CHEMOTAXIS
36
Vol. 60, No. 1
at 1.1.lO-5 M shows a mean CI of 1.56 with a standard deviation of 0.18. It does therefore possess a chemotactic activity on polymorphonuclear leukocytes (CI>l.OO). Hirudin at 1.32.10-s M has a mean CI of 1.03 and standard deviation of 0.13. Hirudin shows a very weak chemotactic chemotactic activity (the CI rests near to 1.00). The cithrombin-hirudin mixture has a mean CI of 0.82 and the standard deviation is 0.12. This mixture is not therefore chemoattractive (the CI lies within the zone of spontaneous mobility). The Student t test shows that there is a significant difference (~(0.05) between the CI's of thrombin and the CI's of the octhrombin-hirudin mixture. Also the CI's obtained for hirudin are statistically different from those obtained for the cxthrombin-hirudin mixture (~(0.05).
CI
18 L6 1,4 1,2 l,O 0.8 0,6 FMLP
Thrombin
HiNdin
Thrombin+Hir Tests
Fiiure 1. Study of the chemotactic activity of FMLP (positive control), of thrombin at 1.1.1O-5 M, of hirudin at 1.32.1O-5 M, of thrombin-hirudin mixture (Thrombin + Hir) (n=ls). The hachured zone shows the area in which all the CI values obt,ainedfor the negative control are comprised. This zone represents the spontaneous mobility of polymorphonuclear leukocytes. +%tatistically different.
*Study of the chemotactic activity of human crthrombin-antithrombin III mixture on polymorphonuclear leukocytes (figure 2) series all of the results obtained can be In this second experimental validated because the positive control gives a mean CI value of 1.62 and a standard deviation of 0.16 and the negative control shows a mean CI of 0.88 and a standard deviation of 0.09. The mean CI obtained for the human
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
37
dthrombin preparation at 1.1.10 -5 M is 1.23 with a standard deviation of 0.08. Thrombin is therefore chemotactic for polymorphonuclear leukocytes. The human antithrombin III at 4.95.10-" M has a mean CI of 1.00 and a standard deviation of 0.07. This CI is very close to 1.00 and shows that antithrombin III does not have a chemoattractive power. The octhrombin-antithrombin III mixture has a mean CI of 0.73 and a. standard deviation of 0.15. It is not chemotactic. Using the Student t test, statistical analysis shows a significant difference between the obtained CI values for orthrombin and antithrombin III? as well as for thrombin and thrombin-antithrombin III mixture (~(0.05).
CI 2.2
3 IT
2.0
198
FMLP
Thrombin
AT
Thrombin+AT Tests
Figure 2. Study of the chemotacticactivity of FMLP of thrombin at 1.1.10 -' M, of antithrombin III (AT III) at 4.95.10-' M, of thrombin inhibited by antithrombin III with a molecular ratio of 1 mol of thrombin for 4.5 mol of antithrombin III (Thrombin + AT) (n=ls).
The hachured zone representsthe spontaneousmobility of polymorphonuclear leukocytes.This area is obtainedwith the help of the CI values obtainedfor the negativecontrol. Statisticallydifferent.
DISCUSSION
Human ~(thrombin attracts human polymorphonuclear leukocytes. It is now etablished that this is equally true for thrombin inactivated at its catalytic site and for 8 thrombin (11). The latter, obtained by autolysis or trypsinolysis of o(thrombin, has a non modified active site, whereas its fibrinogen recognition site is no longer functional (16,17). The structural domain responsible for the chemotactic action of thrombin on polymorphonuclear leukocytes is therefore independent of the catalytic site of the enzyme as well as the fibrinogen recognition site.
38
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
(1)
ANTITHROMBIN Ill
HIRUDIN
(4)
(3) Figure 3.
(1) Spatial representation of human cxthrombin proposed by FENTON II J.W. (26). (2) Representation of thrombin molecule areas which are involved in the interaction with hirudin and antithrombin III. (3) Scheme of the thrombin-hirudin interaction proposed by MARKWART F. (27). x x x Interaction zones between thrombin and hirudin. (4) Scheme of the thrombin-antithrombin III interaction. x x x Interaction zone between thrombin and antithrombin III.
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
39
The hirudin is responsible for the disappearance of the chemotactic power of o(thrombin on polymorphonuclear leukocytes because the thrombin-hirudin complex has no chemotactic action. We can therefore think that hirudin-thrombin interaction masks the structural domain or domains responsible for the chemotactic properties of the enzyme. Hirudin is a specific inhibitor of thrombin, with which it forms a non covalent complex with a 1:l stoechiometry, in this complex the enzyme has lost its clotting activity. Hirudin-thrombin interaction, of the ionic type, is not due to the thrombin catalytic site (3,19,20). The binding of this inhibitor on thrombin happens at two different places on the enzyme B chain. The C terminal segment of hirudin, rich in acid amino acids (21,22) binds to an anionic binding exosite located to the right of the thrombin catalytic site (23,24,25,26,27,28) whereas the N terminal portion of the hirudin molecule interacts with a site close to the catalytic site and localized to i~t'sleft: the apolar binding site(figure 3) (24,25,26,27). This suggests that the chemotactic domain or domains of the thrombin molecule are localized either at the anionic binding exosite of the enzyme or at its apolar binding site, or both. The study carried out with antithrombin III backs up one of these hypotheses. Like the thrombin-hirudin complex ; the thrombin-antithrombin III complex has no chemotactic activity on polymorphonuclear leukocytes. The antithrombin III, by its binding must therefore mask the structures responsible for the chemotaxis performed by the thrombin molecule. The antithrombin III makes a stable and equimolar complex with thrombin. Their association is carried out by the forming of a covalent bond between the Ser of the thrombin catalytic site (chain B of the enzyme) and the Arg 385 close to the C terminal portion of the antithrombin III molecule (15,29) (fig 3). However outside of this covalent bond, hydrophobic bonds are presumed between thrombin and antithrombin III in stabilize the order to complex (2,30,31,32,33). In fact it has been demonstrated that 8 thrombin which is cleaved at the anionic binding exosite, binds as well antithrombin III as octhrombin does (34). Moreover, antithrombin III is able to remove from thrombin the proflavin molecule which is a marker of the apolar binding site localized to the left of the catalytic site of the thrombin molecule (35). As a result, the molecular domain responsible for the attractive action of thrombin on polymorphonuclear leukocytes, seems therefore to be found at the common binding site for hirudin and antithrombin III. This domain is localized to the left of the catalytic site of thrombin, probably at the apolar binding site (fig 3). It is interesting to note as well, that the chemotactic response of monocytes to thrombin is also neutralized by the binding of hirudin and antithrombin III on this enzyme (6,9). The structural domain responsible for the attractive power of thrombin on polymorphonuclear leukocytes and monocytes will therefore be the same.
CONCLUSION Human mthrombin, at the concentration tested is a cytotaxin for polymorphonuclear leukocytes. Its chemotactic action disappears when it is complexed with hirudin or antithrombin III. Hirudin and antithrombin III, once bound to thrombin thus mask the structure of the thrombin molecule responsible for its chemotactic activity. For this enzyme, the common binding site for hirudin and antithrombin III is an apolar binding site localized to the left of the catalytic site. Thus this structure plays a role on the chemotactic properties of the crthrombin with regard to polymorphonuclear leukocytes.
40
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
1.
FENTON II, J.W., FASCO, M.J., STACKROW, A.B., ARONSON, D.L., YOUNG, A.M. and FINLAYSON, J.S. Human thrombins.Production evaluation and properties of cxthrombin. J. Biol. Chem. 252, 3587-3598, 1977.
2.
FENTON II, J.W. Thrombin specificity. Ann. NY Acad. Sci. 370, 468-495, 1981.
3.
FENTON II, J.W. Thrombin. Ann. NY Acad. Sci. 485,
4.
WALZ, D.A., ANDERSON, G.F., CIAGLOWSKI, R.E., AIKEN, M. and FENTON II, J.W. Thrombin elicited contractile responses of aortic smooth muscle (42 211). Proc. Sot. Exp. Biol. Med. 180, 518-526, 1985.
5.
BIZIOS, R., LAI, L., FENTON II, J.W., SONDER, S.A. and MALIK, A.B. Thrombin induced aggregation of lymphocytes non enzymic induction by an hirudin blocked thrombin exosite. Thromb. Res. 38, 425-431, 1985.
6.
BAR SHAVIT, R., KAHN, A., FENTON II, J.W. and WILNER, G.D. Chemotactic response of monocytes to thrombin. J. Cell. Biol. 96, 282-285, 1983.
7.
BAR SHAVIT, R., KAHN, A., FENTON II, J.W. and WILNER, G.D. Receptor mediated chemotactic response of a macrophage cell line (J 776) to thrombin. Lab. Invest. 49, 702-707, 1983.
8.
BAR SHAVIT, R., HRUSKA, K.A., KAHN, A.J. and WILNER, G.D. Hormone like activity of human thrombin. Ann. NY Acad. Sci. 485, 335-348, 1986.
9.
BAR SHAVIT, R., KAHN, A. and WILNER, G.D. Monocyte chemotaxis: stimulation by specific exosite region in thrombin. Science, 220, 728-731, 1983.
10.
BIZIOS, R., LAI, L., FENTON II, J.W. and MALIK, A.B. Thrombin induced neutrophil chemotaxis. Fed. Proc. 43, 973 (Abstr 40261, 1984.
11.
BIZIOS, R., LAI, L., FENTON II, J.W. and MALIK, A.B. Thrombin induced chemotaxis and aggregation of neutrophils. J. Cell. Physiol. 128, 485-490, 1986.
12.
MORIN, A., MARCHAND-ARVIER, M., DOUTREMEPUICH, F. and VIGNERON, C. Coagulation impact on chemotactic activity generation for polymorphonuclear leukocytes. Thromb. Res. (In Press).
13.
CUTLER, J.E. A simple in vitro method for studies on chemotaxis. Proc.
5-15,
1986.
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
41
Sot. Exp. Biol. Med. 147, 471-474, 1974.
14.
MARCHAND-ARVIER, M. and VIGNERON, C. Le chimiotactisme des polynucleaires neutrophiles : moyens d'etude. Feuillets de Biologie, 23, 31-34, 1982.
15.
ROSENBERG, R.D. and DAMU, P.S. The purification and mechanism of action of human antithrombin heparin cofactor. J. Bio:L. Chem. 248, 649o-6505, 1973.
16.
CHANG, J.Y. The structures and proteolytic specificities of autolysed human thrombin. Biochem. J., 240, 797-802, 1986.
17.
ELION, J., BOISSEL, J.P., LE BONNIEC, B., BEZEAUD, A., JANDROT-PERRUS, M RABIET, M.J. and GUILLIN, M.C. Proteolytic derivatives of thrombin. Ann. NY Acad. Sci. 48.5, 16-26, 1986.
18.
MARKWARDT, F. Hirudin as an inhibitor of thrombin. Meth. Enzymol. 19, 924-932,
1970.
19.
DODT, J., KOHLER, S. and BAICI, A. Interaction of site specific hirudin variants with octhrombin. FEBS Lett. 229, 87-90, 1988.
20.
STONE, S.R., DENNIS, S. and HOFSTEENGE, J. Quantitative evaluation of the contribution of ionic interactions to the formation of the thrombin hirudin complex. Biochemistry, 28, 6857-6863, 1989.
21.
KRSTENANSKY, J.L. and MAO, S.J.T. Antithrombin properties of C terminus of hirudin using synthetic unsulfated N O(acety1 hirudin 45-65. FEBS Lett. 211, 10-16, 1987.
22.
MAO, S.J.T., YATES, M.T., OWEN, T.J. and KRSTENANSKY, J.L. Interaction of hirudin with thrombin identification of a minimal binding domain of hirudin that inhibits clotting activity. Biochemistry, 27, 8170-8173, 1988.
23.
KRSTENANSKY, J.L., OWEN, T.J., YATES, M.T. and MAO, S.T.J. Anticoagulant peptides nature of the interaction of the C terminal region of hirudin with a noncatalytic binding site on thrombin. J. Med. Them., 30, 1688-1691,
24.
STONE, S.R., BRAUN, P.J. and HOFSTEENGE, J. Identification of regions of cnthrombin involved in its interaction with h.irudin. Biochemistry, 26, 4617-4624,
25.
1987.
1987.
MARKWARDT, F. Pharmacology of selective thrombin inhibitors. Nouv. Rev.
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
Fr. Hematol. 30, 161-165, 1988.
26.
FENTON II, J.W. Thrombin interactions with Hemost. 15, 265-268, 1989.
27.
MARKWARDT, F. Development of hirudin as an antithrombotic agent. Thromb. Hemost. 15, 269-282, 1989.
28.
CHANG, J.Y. The hirudin binding site of human oithrombin. Identification of lysyl residues with participate in the combining site of hirudin thrombin complex. J. Biol. Chem. 264, 7141-7146, 1989.
29.
FERGUSSON, W.S. and FINLAY, T.H. Formation and stability of the complex formed between human antithrombin III and thrombin. Arch. Biochem. Biophys. 220, 301-308, 1983.
30.
BJORK, I., DANIELSON, A., FENTON II, J.W. and JORNVALL, H. The site in human antithrombin for functional proteolytic cleavage by human thrombin. FEBS Lett. 126, 257-260, 1981.
31.
ROSENBERG, R.D. Chemistry of the hemostatic mechanism and relationship to the action of heparin. Fed. Proc. 36, 10-18, 1977.
32.
BJORK, I. and LINDAHL, U. Mechanism of the anticoagulant action of heparin. Mol. Cell. Biochem. 48, 161-182, 1982.
33.
FENTON II, J.W. and BING, D.H. Thrombin active site regions. Sem. Thromb. Hemost. 12. 200-208. 1986.
34.
CHANG, T.L., FEINMAN, R.D., LANDIS, B.H. and FENTON II, J.W. Antithrombin reactions with o( and 8 thrombins. Biochemistry, 18, 113-119, 1979.
35.
SONDER, S.A. and FENTON II, J.W. Proflavin binding within the fibrinopeptide groove adjacent to the catalytic site of human ot thrombin. Biochemistry, 23, 1818-1823, 1984.
hirudin.
Semin.
Thromb.
its
LOCALIZATIONOF THE STRUCTURALDOMAIN RESPONSIBLEFOR THE CHEMOTACTICPROPERTIES OF THROMBINON POLYMORPHONUCLEAR LEUKOCYTES
A. MORIN, M. MARCHAND ARVIER, F. DOUTREMEPUICH", C. VIGNERON Centre de Transfusion sanguine, Avenue de Bourgogne 54500 Vandoeuvre, France +:-Faculte de Pharmacie, 3 Place de la Victoire 33076 Bordeaux Ckdex, France (Received
2.4.1990;
accepted
(Received
in original form 14.5.1990
by Executive
by Editor O.N. Ulutin)
Editorial Office 3.8.1990)
ABSTRACT Human athrombin at 1.1.1O-5 M is chemotactic for human polymorphonuclear leukoc tes. This thrombin pro erty disappears when the athromr M) hirudin (l.32.lO-. P M) mixture is realized. The same bin (l.l.lO-‘ result is obtained when the thrombin at 1.1. lo-" M is inhibited by antithrombin III in a ratio of 1 mol of thrombin for 4.5 mol of antithrombin III. The hirudin and the antithrombin III appear therefore to mask, by their binding the structural domain responsible for the chemotactic properties of thrombin on polymorphonuclear leukocytes.
INTRODUCTION
Thrombin plays a role of prime importance at the level of the hemostatic system. Its action also extends to plasmatic factors of this system as well as to the vascular and even cellular level (l,2,3). However, outside its action on the hemostatic system tithrombin plays a role in non enzymatic reactions (4,5) in particular it has a chemoattractive power on monocytes and certain macrophage like cell lines (6,7,8). This thrombin property is not due to the catalytic site (6,8,9) nor to the fibrinogen recognition site (7,9) of this enzyme. On the other hand, the thrombin, inhibited by antithrombin III (physiological inhibitor of the enzyme) or by hirudin (specific inhibitor of thrombin) no longer acts like a cytotaxin on monocytes (6,7,9). The inhibitors must therefore mask the site or sites responsible for the chemotaxis carried out by the thrombin on monocytes. It has been shown that oithrombin also exerts a chemotactic activity for the polymorphonuclear leukocytes (10,11,12) but the structural domain of the cxthrombin responsible has not yet been distinguished. The aim of this work is to localize it. To this end the
Key words: Chemotaxis, polymorphonuclear leukocytes, thrombin, antithrombin III, hirudin. 33
34
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
chemotactic activity of the octhrombin will be studied in the presence of inhibitors such as hirudin and antithrombin III, whose interaction sites are known.
MATERIALS ANDMETHOD
*Protocol There are two series of experiments in this work: - Study of the chemotactic activity of human octhrombin-hirudin mixture on polymorphonuclear leukocytes. - Study of the chemotactic activity of human octhrombin-antithrombin III mixture on polymorphonuclear leukocytes. The chemotaxis technique under agarose gel was used after isolation of human leukocytes by accelerated sedimentation on dextran. The experiments were carried out under sterile conditions in avoid bacterial order to contaminations. The chemotactic properties of the compounds belonging to the same experimental series were determined in parallel and with regard to the leukocytes of 5 healthy voluntary subjects. *Reagents Dextran T500 (Pharmacia, Sweden) at 2% in solution NaCl at 9%., NH4Cl at 0.87% in distillated water, glucosed phosphate salin buffer (glucdsed PBS)(137 mM NaCl, 3 mM KCl, 8 mM Na2HC03, 1.4 mM KH2P04, 10 mM glucose, pH = 7.41, agarose (indubiose A37, IBF, France), minimum essential medium (MEM)(concentrated 10 times, Gibco, Scotland), human serum inactivated by heat and conserved at -8ooc, NaHC03 at 7.5%, methanol (Cooperation Pharmaceutique Francaise, France), formaldehyde (Prolabo, France), basic fuschine solution at 1% in ethanol, N formyl-L methionyl-L leucyl-L phenylalanine (FMLP)(Sigma, France) at 10-7-M, human tithrombin (Sigma, France), recombinant hirudin (E.Coli), human antithrombin III (CRTS Nancy, France). *Leukocytes isolation Human blood was collected by veinipuncture in heparin tubes (12 VI/ml of blood). This blood was mixed volume to volume with dextran. After sedimentation the supernatant was collected and then 1 volume of supernatant was mixed with 4 volumes of cold NH4C1(40C) for ten minutes in order to haemolyse the red blood cells. This mixture was centrifuged 8 minutes at 230 g and at 20°C. The leukocyte pellet was washed in glucosed PBS then put into suspension in the same buffer. The number of leukocytes was adjusted at 2.107 polymorphonuclear leukocytes/ml. %hkmot&is under agarose gel The technique used was similar to Cutler's (13) and has already been described by Marchand-Arvier (14). - Gel preparation A gel made from: 1.2% agarose, 79% distillated water, 10% MEM, 10% human serum, 1% NaHC03 was deposited on Petri dishes at a height of 5 mm. - The test In the gel some wells a and b (3 mm diameter, 3 mm distance) were cut. For each chemotactic dish 4 pairs of wells a-b are arranged: 1 pair of wells serves as a negative control, the three others for study of the chemotactic properties of the tested compound. The a wells were filled with 10 pl of the leukocyte suspension whereas b wells were filled with 10 pl of the compound to be tested. The dishes were incubated 3 hours at 37OC under 5% of CO2 in a humid atmosphere. The reaction was stopped by adding 5 ml of methanol for 30 minutes and then 5 ml of formaldehyde (30 minutes equally). The agarose gels were unstuck from the dishes and the leukocytes were visualized using fuschine coloration. Chemotactic mobility (CM) and spontaneous mobility (SM) were
Vol. 60, No. 1
COAGULATION ANDPMNCHEMOTAXIS
35
determined under an optical microscope with a calibrated eye piece. From these results a chemotactic index was calculated : Chemotactic index CI =:!$ - The controls They were carried out with the leukocyte suspension used for the test. . Positive control: performed in 3 pair of wells a-b. Wells b were filled with 10 pl of FMLP whilst wells a were filled with 10 pl of leukocyte suspension. . Negative control: made for each chemotactic dish in a pair of wells a-b. Wells a received 10 ~1 of leukocyte suspension whilst wells b remained empty. The negative control represents the spontaneous mobility of leukocytes. To validate the test the chemotactic mobility (CI) obtained for the positive control must be higher than 1.20 and for the negative control they must be inferior to 1.00. 'ATestedsubstances - CIthrombin and hirudin Human cxthrombin preparation has a biological activity of 3313 NIH/mg. Isoelectrofocalisation of this preparation, thanks to the Phast System (Pharmacia, Sweden) with a gradient of pH from 3 to 9 and after coomassie blue staining only shows a single band. In this study, the final thrombin concentration will be of l.l.lO-5 M as a previous study has shown a chemotactic action of thrombin between (?.28.10e5M and 2.2.10e5 M (12). The thrombin-hirudin mixture is obtained by putting into contact a volume of thrombin at 2.2.10 -5 M in PBS buffer with a volume of hirudin at 2.64.10~" M in the same buffer. The inhibition of thrombin was confirmed: 0.5 ml of EDTA plasma (1.8 mg/ml of blood) added with 10 ~1 of thrombin at 1.1.10m5 M coagulated in 15 seconds whilst 0.5 ml of the same plasma added with 10 ~1 of the thrombin-hirudin mixture had a coagulation time superior to 120 seconds. - ccthrombin and antithrombin III A solution of human athrombin at 1.1.10m5 M was carried out in MEM solution. The preparation of antithrombin III used has a biological activity of 0.94 U/'mg.An isoelectrofocalisation on a pH gradient of 3 to 9 on Phast System of the antithrombin preparation was made. The coomassie blue staining reveals a major band corresponding to antithrombin III and some traces of albumin. The thrombin-antithrombin III complex was obtained by putting into contact 1 mol of thrombin with 4.5 mol of antithrombin III for 30 minutes at 37oC (15). The thrombin inhibition was checked, the coagulation time of 0.5 ml of EDTA plasma added with 10 ~1 of thrombin at 1.1.1O-F'M is 15 seconds whereas the coagulation time of 0.5 ml of the same plasma mixed to 10 f11 of the thrombin-antithrombin III mixture was superior to 12@ seconds. %Statistical analysis For each substance tested, 15 CI values were obtained. The different compounds from the same experimental series were compared with each other using the Student t test. p inferior to 0.05 was considered as significant.
RESULTS
*Study of the chemotactic activity of human o~thrombin-hirudin mixture on polymorphonuclear leukocytes (figure 1) For this exoerimental series the oositive control gives a mean CI of 1.88 and a standard deviation of 0.27 whereas the mean CI value of the negative control (or spontaneous mobility) is 0.88 with a standard deviation of 0.08. So the other results obtained in this series could be validated. The human octhrombin
COAGULATION AND PMN CHEMOTAXIS
36
Vol. 60, No. 1
at 1.1.lO-5 M shows a mean CI of 1.56 with a standard deviation of 0.18. It does therefore possess a chemotactic activity on polymorphonuclear leukocytes (CI>l.OO). Hirudin at 1.32.10-s M has a mean CI of 1.03 and standard deviation of 0.13. Hirudin shows a very weak chemotactic chemotactic activity (the CI rests near to 1.00). The cithrombin-hirudin mixture has a mean CI of 0.82 and the standard deviation is 0.12. This mixture is not therefore chemoattractive (the CI lies within the zone of spontaneous mobility). The Student t test shows that there is a significant difference (~(0.05) between the CI's of thrombin and the CI's of the octhrombin-hirudin mixture. Also the CI's obtained for hirudin are statistically different from those obtained for the cxthrombin-hirudin mixture (~(0.05).
CI
18 L6 1,4 1,2 l,O 0.8 0,6 FMLP
Thrombin
HiNdin
Thrombin+Hir Tests
Fiiure 1. Study of the chemotactic activity of FMLP (positive control), of thrombin at 1.1.1O-5 M, of hirudin at 1.32.1O-5 M, of thrombin-hirudin mixture (Thrombin + Hir) (n=ls). The hachured zone shows the area in which all the CI values obt,ainedfor the negative control are comprised. This zone represents the spontaneous mobility of polymorphonuclear leukocytes. +%tatistically different.
*Study of the chemotactic activity of human crthrombin-antithrombin III mixture on polymorphonuclear leukocytes (figure 2) series all of the results obtained can be In this second experimental validated because the positive control gives a mean CI value of 1.62 and a standard deviation of 0.16 and the negative control shows a mean CI of 0.88 and a standard deviation of 0.09. The mean CI obtained for the human
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
37
dthrombin preparation at 1.1.10 -5 M is 1.23 with a standard deviation of 0.08. Thrombin is therefore chemotactic for polymorphonuclear leukocytes. The human antithrombin III at 4.95.10-" M has a mean CI of 1.00 and a standard deviation of 0.07. This CI is very close to 1.00 and shows that antithrombin III does not have a chemoattractive power. The octhrombin-antithrombin III mixture has a mean CI of 0.73 and a. standard deviation of 0.15. It is not chemotactic. Using the Student t test, statistical analysis shows a significant difference between the obtained CI values for orthrombin and antithrombin III? as well as for thrombin and thrombin-antithrombin III mixture (~(0.05).
CI 2.2
3 IT
2.0
198
FMLP
Thrombin
AT
Thrombin+AT Tests
Figure 2. Study of the chemotacticactivity of FMLP of thrombin at 1.1.10 -' M, of antithrombin III (AT III) at 4.95.10-' M, of thrombin inhibited by antithrombin III with a molecular ratio of 1 mol of thrombin for 4.5 mol of antithrombin III (Thrombin + AT) (n=ls).
The hachured zone representsthe spontaneousmobility of polymorphonuclear leukocytes.This area is obtainedwith the help of the CI values obtainedfor the negativecontrol. Statisticallydifferent.
DISCUSSION
Human ~(thrombin attracts human polymorphonuclear leukocytes. It is now etablished that this is equally true for thrombin inactivated at its catalytic site and for 8 thrombin (11). The latter, obtained by autolysis or trypsinolysis of o(thrombin, has a non modified active site, whereas its fibrinogen recognition site is no longer functional (16,17). The structural domain responsible for the chemotactic action of thrombin on polymorphonuclear leukocytes is therefore independent of the catalytic site of the enzyme as well as the fibrinogen recognition site.
38
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
(1)
ANTITHROMBIN Ill
HIRUDIN
(4)
(3) Figure 3.
(1) Spatial representation of human cxthrombin proposed by FENTON II J.W. (26). (2) Representation of thrombin molecule areas which are involved in the interaction with hirudin and antithrombin III. (3) Scheme of the thrombin-hirudin interaction proposed by MARKWART F. (27). x x x Interaction zones between thrombin and hirudin. (4) Scheme of the thrombin-antithrombin III interaction. x x x Interaction zone between thrombin and antithrombin III.
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
39
The hirudin is responsible for the disappearance of the chemotactic power of o(thrombin on polymorphonuclear leukocytes because the thrombin-hirudin complex has no chemotactic action. We can therefore think that hirudin-thrombin interaction masks the structural domain or domains responsible for the chemotactic properties of the enzyme. Hirudin is a specific inhibitor of thrombin, with which it forms a non covalent complex with a 1:l stoechiometry, in this complex the enzyme has lost its clotting activity. Hirudin-thrombin interaction, of the ionic type, is not due to the thrombin catalytic site (3,19,20). The binding of this inhibitor on thrombin happens at two different places on the enzyme B chain. The C terminal segment of hirudin, rich in acid amino acids (21,22) binds to an anionic binding exosite located to the right of the thrombin catalytic site (23,24,25,26,27,28) whereas the N terminal portion of the hirudin molecule interacts with a site close to the catalytic site and localized to i~t'sleft: the apolar binding site(figure 3) (24,25,26,27). This suggests that the chemotactic domain or domains of the thrombin molecule are localized either at the anionic binding exosite of the enzyme or at its apolar binding site, or both. The study carried out with antithrombin III backs up one of these hypotheses. Like the thrombin-hirudin complex ; the thrombin-antithrombin III complex has no chemotactic activity on polymorphonuclear leukocytes. The antithrombin III, by its binding must therefore mask the structures responsible for the chemotaxis performed by the thrombin molecule. The antithrombin III makes a stable and equimolar complex with thrombin. Their association is carried out by the forming of a covalent bond between the Ser of the thrombin catalytic site (chain B of the enzyme) and the Arg 385 close to the C terminal portion of the antithrombin III molecule (15,29) (fig 3). However outside of this covalent bond, hydrophobic bonds are presumed between thrombin and antithrombin III in stabilize the order to complex (2,30,31,32,33). In fact it has been demonstrated that 8 thrombin which is cleaved at the anionic binding exosite, binds as well antithrombin III as octhrombin does (34). Moreover, antithrombin III is able to remove from thrombin the proflavin molecule which is a marker of the apolar binding site localized to the left of the catalytic site of the thrombin molecule (35). As a result, the molecular domain responsible for the attractive action of thrombin on polymorphonuclear leukocytes, seems therefore to be found at the common binding site for hirudin and antithrombin III. This domain is localized to the left of the catalytic site of thrombin, probably at the apolar binding site (fig 3). It is interesting to note as well, that the chemotactic response of monocytes to thrombin is also neutralized by the binding of hirudin and antithrombin III on this enzyme (6,9). The structural domain responsible for the attractive power of thrombin on polymorphonuclear leukocytes and monocytes will therefore be the same.
CONCLUSION Human mthrombin, at the concentration tested is a cytotaxin for polymorphonuclear leukocytes. Its chemotactic action disappears when it is complexed with hirudin or antithrombin III. Hirudin and antithrombin III, once bound to thrombin thus mask the structure of the thrombin molecule responsible for its chemotactic activity. For this enzyme, the common binding site for hirudin and antithrombin III is an apolar binding site localized to the left of the catalytic site. Thus this structure plays a role on the chemotactic properties of the crthrombin with regard to polymorphonuclear leukocytes.
40
Vol. 60, No. 1
COAGULATION AND PMN CHEMOTAXIS
1.
FENTON II, J.W., FASCO, M.J., STACKROW, A.B., ARONSON, D.L., YOUNG, A.M. and FINLAYSON, J.S. Human thrombins.Production evaluation and properties of cxthrombin. J. Biol. Chem. 252, 3587-3598, 1977.
2.
FENTON II, J.W. Thrombin specificity. Ann. NY Acad. Sci. 370, 468-495, 1981.
3.
FENTON II, J.W. Thrombin. Ann. NY Acad. Sci. 485,
4.
WALZ, D.A., ANDERSON, G.F., CIAGLOWSKI, R.E., AIKEN, M. and FENTON II, J.W. Thrombin elicited contractile responses of aortic smooth muscle (42 211). Proc. Sot. Exp. Biol. Med. 180, 518-526, 1985.
5.
BIZIOS, R., LAI, L., FENTON II, J.W., SONDER, S.A. and MALIK, A.B. Thrombin induced aggregation of lymphocytes non enzymic induction by an hirudin blocked thrombin exosite. Thromb. Res. 38, 425-431, 1985.
6.
BAR SHAVIT, R., KAHN, A., FENTON II, J.W. and WILNER, G.D. Chemotactic response of monocytes to thrombin. J. Cell. Biol. 96, 282-285, 1983.
7.
BAR SHAVIT, R., KAHN, A., FENTON II, J.W. and WILNER, G.D. Receptor mediated chemotactic response of a macrophage cell line (J 776) to thrombin. Lab. Invest. 49, 702-707, 1983.
8.
BAR SHAVIT, R., HRUSKA, K.A., KAHN, A.J. and WILNER, G.D. Hormone like activity of human thrombin. Ann. NY Acad. Sci. 485, 335-348, 1986.
9.
BAR SHAVIT, R., KAHN, A. and WILNER, G.D. Monocyte chemotaxis: stimulation by specific exosite region in thrombin. Science, 220, 728-731, 1983.
10.
BIZIOS, R., LAI, L., FENTON II, J.W. and MALIK, A.B. Thrombin induced neutrophil chemotaxis. Fed. Proc. 43, 973 (Abstr 40261, 1984.
11.
BIZIOS, R., LAI, L., FENTON II, J.W. and MALIK, A.B. Thrombin induced chemotaxis and aggregation of neutrophils. J. Cell. Physiol. 128, 485-490, 1986.
12.
MORIN, A., MARCHAND-ARVIER, M., DOUTREMEPUICH, F. and VIGNERON, C. Coagulation impact on chemotactic activity generation for polymorphonuclear leukocytes. Thromb. Res. (In Press).
13.
CUTLER, J.E. A simple in vitro method for studies on chemotaxis. Proc.
5-15,
1986.
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
41
Sot. Exp. Biol. Med. 147, 471-474, 1974.
14.
MARCHAND-ARVIER, M. and VIGNERON, C. Le chimiotactisme des polynucleaires neutrophiles : moyens d'etude. Feuillets de Biologie, 23, 31-34, 1982.
15.
ROSENBERG, R.D. and DAMU, P.S. The purification and mechanism of action of human antithrombin heparin cofactor. J. Bio:L. Chem. 248, 649o-6505, 1973.
16.
CHANG, J.Y. The structures and proteolytic specificities of autolysed human thrombin. Biochem. J., 240, 797-802, 1986.
17.
ELION, J., BOISSEL, J.P., LE BONNIEC, B., BEZEAUD, A., JANDROT-PERRUS, M RABIET, M.J. and GUILLIN, M.C. Proteolytic derivatives of thrombin. Ann. NY Acad. Sci. 48.5, 16-26, 1986.
18.
MARKWARDT, F. Hirudin as an inhibitor of thrombin. Meth. Enzymol. 19, 924-932,
1970.
19.
DODT, J., KOHLER, S. and BAICI, A. Interaction of site specific hirudin variants with octhrombin. FEBS Lett. 229, 87-90, 1988.
20.
STONE, S.R., DENNIS, S. and HOFSTEENGE, J. Quantitative evaluation of the contribution of ionic interactions to the formation of the thrombin hirudin complex. Biochemistry, 28, 6857-6863, 1989.
21.
KRSTENANSKY, J.L. and MAO, S.J.T. Antithrombin properties of C terminus of hirudin using synthetic unsulfated N O(acety1 hirudin 45-65. FEBS Lett. 211, 10-16, 1987.
22.
MAO, S.J.T., YATES, M.T., OWEN, T.J. and KRSTENANSKY, J.L. Interaction of hirudin with thrombin identification of a minimal binding domain of hirudin that inhibits clotting activity. Biochemistry, 27, 8170-8173, 1988.
23.
KRSTENANSKY, J.L., OWEN, T.J., YATES, M.T. and MAO, S.T.J. Anticoagulant peptides nature of the interaction of the C terminal region of hirudin with a noncatalytic binding site on thrombin. J. Med. Them., 30, 1688-1691,
24.
STONE, S.R., BRAUN, P.J. and HOFSTEENGE, J. Identification of regions of cnthrombin involved in its interaction with h.irudin. Biochemistry, 26, 4617-4624,
25.
1987.
1987.
MARKWARDT, F. Pharmacology of selective thrombin inhibitors. Nouv. Rev.
COAGULATION AND PMN CHEMOTAXIS
Vol. 60, No. 1
Fr. Hematol. 30, 161-165, 1988.
26.
FENTON II, J.W. Thrombin interactions with Hemost. 15, 265-268, 1989.
27.
MARKWARDT, F. Development of hirudin as an antithrombotic agent. Thromb. Hemost. 15, 269-282, 1989.
28.
CHANG, J.Y. The hirudin binding site of human oithrombin. Identification of lysyl residues with participate in the combining site of hirudin thrombin complex. J. Biol. Chem. 264, 7141-7146, 1989.
29.
FERGUSSON, W.S. and FINLAY, T.H. Formation and stability of the complex formed between human antithrombin III and thrombin. Arch. Biochem. Biophys. 220, 301-308, 1983.
30.
BJORK, I., DANIELSON, A., FENTON II, J.W. and JORNVALL, H. The site in human antithrombin for functional proteolytic cleavage by human thrombin. FEBS Lett. 126, 257-260, 1981.
31.
ROSENBERG, R.D. Chemistry of the hemostatic mechanism and relationship to the action of heparin. Fed. Proc. 36, 10-18, 1977.
32.
BJORK, I. and LINDAHL, U. Mechanism of the anticoagulant action of heparin. Mol. Cell. Biochem. 48, 161-182, 1982.
33.
FENTON II, J.W. and BING, D.H. Thrombin active site regions. Sem. Thromb. Hemost. 12. 200-208. 1986.
34.
CHANG, T.L., FEINMAN, R.D., LANDIS, B.H. and FENTON II, J.W. Antithrombin reactions with o( and 8 thrombins. Biochemistry, 18, 113-119, 1979.
35.
SONDER, S.A. and FENTON II, J.W. Proflavin binding within the fibrinopeptide groove adjacent to the catalytic site of human ot thrombin. Biochemistry, 23, 1818-1823, 1984.
hirudin.
Semin.
Thromb.
its
