- Email: [email protected]
IIIa complex
THROMBOSIS RESEARCH 69; 43!5-442,1993 0049-3848/93 $6.00 + .OOPrinted in the USA. Copyright (c) 1993 Pergamon Press Ltd. All rights reserved.
FUNCTIONAL PLATELET
STUDIES FUNCTION
ON PLATELETS OF A PATIENT WITH AN ACQUIRED DISORDER ASSOCIATED WITH AUTOANTIBODIES AGAINST MEMBRANE GLYCOPROTEIN IIB/IIIA COMPLEX
OF
P. Spangenberg’, C.M. Kirchmaier’, A. Schirmer’, M. Meyer3 and H.K. Breddin’ Institute of Pathological Biochemistry’ and Department of Genetics3 of the School of Medicine of Erfurt and Department of Angiology’ of the Center for Internal Medicine, J.W. Goethe University Frankfurt/M., F.R.G.
(Received 3.10.1992; accepted in revised form 23.11.1992 by Editor S. Lopaciuk) (Received by Executive Editorial Office 3.1.1993)
Abstract Platelet functions have been studied of a 63 year old woman with a severe acquired thrombopathy.The platelets did not adhere to siliconized glass. Aggregation could not be induced by either ADP (1 /IM) nor collagen (2 flglml), no release of serotonin was found under these conditions. Thrombin caused only a weak aggregation response. Quantitative analvsis of platelet actin revealed a very low total actin content (473 /./g/10’ platelets) .and an extremely low F-actin value 13% of total actin). Stimulation of platelets with 0.1 U/ml thrombin for 3 min resulted in an increase of only 5% ,F; actin, whereas ADP and collagen did not induce any actin polymerization. Ca movement in the patient’s platelets is severely impaired after ADP and collagen stimulation, whereas a normal Ca” movement was obtained by 0.1 U/ml thrombin. The inhibition of the functions of normal platelets (aggregation and actin polymerization) by addition of patient’s serum (5-10% final concentration) points to receptor blockade by platelet autoantibodies in the patient’s serum. The antibody was purified by adsorption on Protein-A-Sepharose. Addition of IgG-suspension (5% final concentration) to washed control platelets resulted in similar effects on aggregation and actin polymerization compared to the effects of patient’s serum.
Autoantibodies chronic against surface antigens are responsible for autoimmune thrombocytopenic purpura (l-3). The increased peripheral platelet destruction found in this disease is due to the cytotoxic effect of the autoantibodies toward the platelets. Frequently the antigenic determinants are associated with the glycoproteins Ilb/llla but the exact antigenic specificity is as Yet unknown. In autoimmune thrombocytopenic purpura also autoantibodies against glycoprotein lb were found (4). Until now they were underscored, but using a more sensitive method a recent study demonstrated the same frequency of glycoprotein lb autoantibodies as those against glycoproteins llblllla (5). The patient reported here shows a platelet function defect caused by autoantibodies towards the glycoprotein Ilb/llla complex without immunoinjury of the platelets (6). The patient exhibits a severe bleeding tendency which developed during life. We have investigated the functional behaviour of the patient’s platelets and the effects of patient’s serum and isolated IgG from Key words: platelets, autoantibodies, aggregation, calcium, actin, platelet dysfunction Corresponding author: P. Spangenberg, Institute of Pathological Biochemistry, School of Medicine of Erfurt, Nordhauser Str. 74, D-0-5010 Erfurt, F.R.G. 435
PLATELET AUTOANTIBODIES
patient’s
serum on the function
Vol. 69, No. 5
of platelets from healthy volunteers.
MATERIALS
AND METHODS
Blood was obtained from the patient and control subjects by venipuncture anticoagulated with 3.13% trisodium citrate (9 parts of blood and 1 part of trisodium Platelet rich plasma (PRP) was obtained by centrifugation platelets were prepared as previously described (7).
at 200
x Q for
15
min.
AQQreQatiOt'Iwas measured turbidometrically according to Born (8). Concentrations used in this study were 1 ,vM (ADP), 2 /./g/ml (collagen) and 0.1 U/ml (thrombin). The release reaction was measured by the amount of “C-5-hydroxytryptamine from the platelets that had been pre-labelled with this agent (9).
and was citrate). Washed
of agonists
(5HT)
released
Actin determinations were performed by the DNase I inhibition assay (10) in the modification Holme et al. (11) as previously described (12).
of
Calcium movements were followed measurements of INDO 1 /AM.
by
fluorometrically
according
to
Schirmer
et
al.
(7)
IgG from patient’s serum was isolated using a Protein-A-Sepharose column (Pharmacia, Sweden). The acetic acid-desorbed material was neutralized immediately and used in the replacement experiments.
RESULTS Case report. a 62 year-old female, was admitted to hospital in 1982 because of The patient Sch., paraproteinemia. In the same year unclear blood losses started and under the diagnoses of autoimmunohaemolytic anemia she was repeatedly treated with prednisolon and blood transfusions. In 1984 the haemorrhagic diathesis persisted with severe gastrointestinal bleeding In 1985 the diagnosis of idiopathic and epistaxis necessitating multiple blood transfusions. thrombocytopenic purpura was made and the patient was splenectomized. The low platelet count increased to subnormal and normal values, but the bleeding tendency persisted. In 1987 haemostatic tests indicated a normal platelet count, prolonged bleeding time, strongly reduced platelet adhesion and defective platelet aggregation. Aaareaation studies. Pig. 1 reports platelet aggregation responses obtained with patient’s PRP or washed platelets. Platelets did not aggregate with neither ADP (1 ,uM) nor collagen (2 pug/ml) and did not release considerable amounts of 5HT (4% release in ADP-induced platelets as compared to 16% release in control platelets using 1 PM ADP, 0% release in collagen-induced platelets as compared to 42% release in control platelets induced by 2 ,ug/ml collagen). Thrombin (0.1 U/ml) caused only a slight aggregation response compared to the thrombin-induced aggregation in control washed platelets. The platelets did not adhere to siliconized glass (index = 0.0, normal value 0.8 - 1.4). Patient’s serum induced a similar defect when mixed to control washed platelets. This effect was concentration-dependent and a complete inhibition by patient’s serum in case of ADP- or collagen-induced platelet aggregation was achieved at a concentration of 5% serum ((v/v), see Pig. 1). A similar effect was observed when patient’s serum was mixed to control washed This effect was also concentration-dependent, but only platelets and stimulated with thrombin. a partial inhibition resulted even in case of 5% patient’s serum ((v/v), see Pig. 1). A very similar behaviour was obtained using isolated IgG at a concentration of 5% (v/v), ADP- and collageninduced aggregation were completely blocked (data not shown here).
Vol. 69, No. 5
ADP
‘RP
PLATELET AUTOANTIBODIES
, collagen 7
I Irl------thrombin
HP
1
437
control
thrombin
WP
collagen
I
I
ADP
I
I
1:: patient
1
1
1
1
2 control
2
2 1
: patient’s
2
: control
serum serum
added added
FIG. 1. Aggregation of platelets from the patient and control subjects in the presence of patient’s serum. Patient’s platelets in plasma (PRP) were stimulated with ADP (1 PM) or collagen (2 /Ig/ml), washed patient’s platelets (WP) were stimulated with thrombin (0.1 U/ml). In the replacement experiments control washed platelets (WP) in the presence of control or patient’s serum (!5% final concentration) were subjected to thrombin (0.1 U/ml), collagen (2 pQ/ti) and ADP (I ,YM).
Calcium movements. As shown in Fio. 2 oatient’s platelets loaded with INDO l/AM and activated with thrombin or ionomycin exhibited ‘a quite normal increase of cytosolic Ca”, whereas those stimulated with ADP or collagen did not show any calcium signal. In the replacement experiments (control platelets in the presence of patient’s serum) this effect on the calcium movement was not obtained. ADP as well as collagen stimulated the typical Ca *’ increase in the cytoplasma as found in the case of control platelets in the presence of control serum (see Fig. 2, lower Part). Actin determinations. In the platelets of the patients extremely low value for F-actin normally towards stimulation lasted for 1 min, the cells were
a significantly smaller total actin content was measured and an was found (see Table I). These platelets did not polymerize actin with ADP, collagen or thrombin (stimulation with the agonists then lyzed and the actin was measured). ADP and collagen were
438
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
not able to induce any actin polymerization, whereas thrombin caused a polymerization of only 5% (the increase of F-actin found normally after thrombin-stimulation is about 30% of total actin).
+
tca2+1 - rca2fext - ca2f ext ext
+ tca2fext
rca2++ r&l
thrombin
1
300250-
200-
ADP
ionomycin
1
1
1
r-_
7,
‘!-T
0;
thrombin
collagen
5
0
T-7
1
0
1
min
collagen
_
nM
ADP
FIG. 2. Fluorescence tracings on activation of INDO 1 /AM-loaded platelets from the patient (upper part) and from a normal subject (lower part) in the absence (1) or presence (2) of patient’s serum (5% final concentration) with thrombin (0.1 U/ml), ADP (1 ,uM), collagen (2 pg/ml) or ionomycin (1 MM).
Actin polymerization was also studied in control washed platelets in the presence of patient’s serum (5% final concentration, v/v). As shown in Table II patient’s serum inhibits completely the whereas a weak actin polymerization is ADPand collagen-induced actin polymerization, detectable in thrombin-stimulated platelets. A similar behaviour is obtained when lgG (5% final concentration, v/v) is added to control platelets (data not shown here).
DISCUSSION As discussed in a previous paper (6) the platelet defect of the patient investigated here can be characterized as an acquired thrombopathy due to autoantibodies against the membrane glycoprotein Ilb/llla complex. The autoantibodies do not cause an increased peripheral platelet
Vol. 69, No. 5
PLATELET AUTOANTIBODIES
destruction, as it occurs in patients with chronic idiopathic thrombocytopenic shown by Meyer et al. (6) the autoantibodies are IgG-type antibodies.
TABLE Actin Analysis
total actin (pg 110’ platelets)
controls (n = 50)
patient
991
473
f
148
+ 12.3
97
F-actin (% of total actin)
20.2
+ 12.3
3
due to stimulation
ADP (1 PM) COLLAGEN (2 pglml) THROMBIN (0.1 U/ml)
As
of Platelets of the Patient and of Control Subjects
79.8
polymerization
(l-3).
I
G-actin (% of total actin)
Actin
purpura
of platelets
increase in F-actin (in % of total actin) controls (n = 5)
increase in F-actin (in % of total actin) patient
32 + 17 42 f 5 30-+ 1
0 0 5
Splenectomy of the patient who has shown originally an idiopathic thrombocytopenic purpura (6) induced a normalization of the platelet count (normal to subnormal platelet count), but haemorrhagic symptoms did not disappear. As described by Niessner et al. (13) and Balduini et al. (I 4) autoantibodies against the membrane glycoprotein llblllla complex with different complex with different antigenic determinants lead either to immunoinjury of platelets or to platelet function defects. Aggregation studies revealed a defective response to ADP, collagen and thrombin. Balduini et al. (14) reported a very similar case of a woman with autoantibodies against the glycoprotein Ilb/llla complex. Moreover, the patient’s serum induced the same defect when mixed with normal platelets, indicating that the autoantibody is present in serum and that the autoantibody is able to block the receptors of control platelets. The actin analysis of the patient’s platelets revealed atypical results. The total platelet actin content (473 ,eg/lO’ platelets) is less than half of the amount found in normal control platelets (991 + 148 pg/lO’ platelets). Whether this is the result of a defective actin equipment of platelets during thrombopoiesis or whether platelets possess a normal actin content which is not detectable due to the presence of actin-protein-complexes (actin-capping), which are still available in the platelet lysate and which may fail to inhibit the DNase I (biochemical assay of actin determination) remains open. The F-actin content of the platelets (3% of total platelet actin) is extremely low, since 20.2 + 12.1 % F-actin of total actin for females older than 35 years have been determined recently (12).
440
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
TABLE II Effect of Patient’s
Actin
Serum on Agonist-Induced
polymerization
due to stimulation
ADP (1 /JM) COLLAGEN (2 I.rglml) THROMBIN (0.1 U/ml)
Actin Polymerization
in Control
Platelets
of platelets
increase in F-actin (in % of total actin)
increase in F-actin (in % of total actin)
control platelets + control serum (n = 3)
control platelets + patient serum
18 * 27+ 30+
0 0 5
11 8 1
The F-actin content has been considered to be a parameter of platelet reactivity and donors with a high F-actin content (e.g. diabetics, about 40% F-actin) have been characterized to possess hyperreactive platelets (12). Consequently, the very low F-actin content of the patient investigated here is in correspondence to the defective aggregation behaviour. Physiological agonists rapidly induce a conversion of G- into F-actin (actin polymerization); e.g. thrombin increases within 15 set the F-actin content from 40 to 60-80 % (15-18). These experiments have been performed with washed platelets, which become preactivated during the preparation leading to increased basal F-actin values (12). Actin polymerization induced by ADP or collagen can also be studied in PRP and the increase in F-actin is agonist-dependent (see Table I, platelets of healthy volunteers). The autoantibodies against the Qlycoprotein Ilb/llla of the patient inhibit completely the ADP- and collagen-induced actin polymerization in the patient’s platelets, the thrombin response is very small (5% F-actin increase versus 30% in control platelets). Interestingly, the same effects occur when control platelets are mixed with patient’s serum. The ADP- and collagen-responses are completely inhibited again, whereas only 5% increase in F-actin was measurable when the cells were stimulated by thrombin. Since washed platelets were mixed with control serum and used for these experiments the increase in F-actin is smaller than the F-actin is smaller than the increase found in PRP during corresponding activation due to the prea.ctivation of platelets during the preparation (Table I versus Table II). The regulation of actin polymerization in platelets is still an open problem. The process seems to be receptor-mediated and the inhibitory effects of the llblllla autoantibodies on normal platelets described here support this statement. Moreover, transducer or effector systems might be involved in the regulation of actin polymerization (19, 20). We have shown recently that proteinkinase C and ,Ca” ions are somehow involved in the reorganization of microfjdaments in platelets and in the agonist-induced actin polymerization (21-23). Previous experiments Iwith phorbol ester and ionophore A 23187 failed to define a role for these intracellular mediators II 9, 20). The calcium movement experiments performed on the patient’s platelets would fit with the concept that calcium ions are involved in the regulation of actin polymerization since ADP- and collagen-induced increase of cytosolic Ca ‘+ is inhibited in those platelets (coincidental with the inhibition of aggregation and actin polymerization). The interpretation of the thrombin response however is more difficult since both aggregation and actin polymerization are inhibited but a normal calcium response is observed. Also the replacement experiments seem not to be congruent with the idea of the involvement of calcium ions in actin polymerization. The ADP-and collagen-induced calcium responses are quite normal in control platelets mixed with patient’s serum, although aggregation and actin polymerization are still completely inhibited. One possible
Vol. 69. No. 5
441
PLATELET AUTOANTIBODIES
explanation for this could be an internalization of the autoantibodies which may occur during the thrombopoiesis and which may inhibit the Ca” movement usually induced by collagen and ADP, whereas the autoantibodies in the replacement experiments do not become internalized (incubation times up to 1 h) and are not able to inhibit the Ca” movement. Although the F-actin content is very low in the platelets of the patient and actin polymerization is impaired the clot retraction is still normal (6). For clot retraction a structural coupling between actin, myosin and the fibrin network is essential. Van Deurs and Behnke (24) reported the existence of intramembrane particles which were evenly distributed in nonactivated platelets. The distribution of these particles did not change during platelet activation and aggregation (where actin polymerization occurs), but clusters of particles were found during clot retraction. It was suggested that this redistribution of the intramembrane particles reflects the mechanical transmembrane links between the actinlmyosin and the fibrin net. In conclusion, the Ilb/llla autoantibodies found in the patient described here cause dramatic defects in platelet function and mimic a similar situation when mixed with control platelets. These autoantibodies are therefore an interesting tool to study mechanisms of platelet activation.
Acknowledgements This work was supported by a grant from the International Institute of Thrombosis Diseases Frankfurt/M., F.R.G. and by a grant from the DFG No. llB5-Ki429/1-1
and Vascular
REFERENCES I. VAN LEEUWEN, E.F., VAN DER VEN, J.Th.M., ENGELFRIET, C.P. and VON DEM BORNE, A.E.G.Kr. Specificity of autoantibodies in autoimmune thrombocytopenenic purpura. Blood, 59, 23-26, 1982. 2. VARON, D. and KARPATKIN, S. A monoclonal anti-platelet antibody with decreased reactivity for autoimmune thrombocytopenic platelets. Proc. Natl. Acad. Sci. USA, 80, 69926998, 1983. 3. WOODS, V.L., OH, E.H., MASON, D. and MCMILLAN, R. Autoantibodies against the platelet glycoprotein Ilb/llla complex in patients with chronic ITP. Blood 63, 368-375, 1984. 4. WOODS, V.L., KURATA, Y., MONTGOMERY, R.R., TANI; P., MASON; D., OH, E.H. and MCMILLAN, R. Autoantibodies against platelet glycoprotein lb in patients with chronic immune thrombocytopenic purpura. Blood, 64, 156-l 60, 1984. 5. KIEFEL, V., SANTOSO, S., KAUFMANN, E. and MUELLER-ECKHARDT, C. Autoantibodies against glycoprotein lb/IX: a frequent finding in autoimmune thrombocytopenic purpura. Br. J. Haematol. 79, 256-262, 1991. 6. MEYER, M., KIRCHMAIER, CM., SCHIRMER, A., SPANGENBERG, P., STRGHL, Ch. and BREDDIN, H.K. Acquired thrombopathy: severe disorder of platelet function associated with autoantibodies against membrane glycoprotein Ilb-llla complex. 1. Glycoprotein analysis. Thrombos. Haemost., 65, 491-496, 1991. 7. SCHIRMER, A., BELLINGER, O.K. and KIRCHMAIER, C.M. Freies intrazytoplasmatisches Kalzium in Thrombozyten bei Gesunden und Patienten mit Thrombopathien. Inn. Med., 15, 151155, 1988. 8. BORN, G.V.R. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature, 194, 927-929, 1962. 9. HEPTINSTALL, S., BEAVEN, J., COCKBILL, S.R., HANLEY, S.P. and PARRY, M.J. Effects of a selective inhibitor of thromboxane svnthetase on human blood platelet behaviour. Thrombos. Res., 20, 219-230, 1980. 10. BLIKSTAD, I., MARKEY, F., CARLSSON, L., PERSSON, S. and LINDBERG, U. Selective assay of monomeric and filamentous actin in cell extracts. Cell, 15, 935-943, 1978. 11. HOLME, T., KELLIE, S., WYKE, J.A. and CRAWFORD, N. Effect of transforamtion by Rous sarcoma virus on the character and distribution of actin in rat-l fibroblasts: a biochemical and microscopical study. 8r. J. Cancer, 53, 465-476, 1986.
442
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
12. SPANGENBERG, P. and TILL, U. Actin filament content in platelets - a sensitive index of cellular reactivity. Biosc. Rep., 9, 307-313, 1989. 13. NIESSNER, H., CLEMETSON, K.J., PANZER, S., MUELLER-ECKHARDT, Ch., SANTOSO, S. and BETTELHEIM, P. Acquired thrombasthenia due to GP Ilb/llla-specific autoantibodies. Blood, 68, 571-576, 1986. 14. BALDUINI, C.L., GRIGNANI, G., SINIGAGLIA, F., BISIO, A., PACCHIARINI, L., ROTA SCALABRINI, D., BALDUINI, C., MAURI, C. and ASCARI, E. Severe platelet dysfunction in a patient with autoantibodies against membrane glycoprotein Ilb/llla. Haemost., 7, 98-l 04, 1987. 15. CARLSSON, L., MARKEY, F., BLIKSTAD, I., PERSSON, T. and LINDBERG; U. Reorganization of actin in platelets stimulated by thrombin as measured by the DNase I inhibition assay. Proc. Natl. Acad. Sci. USA, 76, 6376-6380, 1979. 16. FOX, E.J.B. and PHILLIPS, D.R., Inhibition of actin polymerization in blood platelets by cytochalasins. Nature, 292, 650-652, 1981. 17. CASELLA, J.F., FLANAGAN, M.D. and LIN, S. Cytochalasin D inhibits actin polymerization and induces depolymerization of actin filaments formed during platelet shape change. Nature, 293, 302-305, 1981. 18. PRIBLUDA, V. and ROTMAN, A. Dynamics of membrane-cytoskeleton interaction in activated blood platelets. Biochem., 2 1, 2825-2832, 1982. 19. LASSING, I. and LINDBERG, U. Specific interaction between phosphatidylinositol-4,5bisphosphate and profilactin. Nature, 314, 473-474, 1985. 20. LASSING, I. and LINDBERG, U. Evidence that the phosphatidylinositol cycle is linked to cell motility. Exp. Cell Res., 174, l-l 5, 1988. 21. SPANGENBERG, P., KOTELIANSKY, V.E., LUKASHEV, M.E. and TILL, U. EinfluB von Phorbolester (PMA) und Forskolin auf das G-/F-Aktin Gleichgewicht humaner Thrombozyten. Acta histochem., 39, 397-402, 1990. 22. SPANGENBERG, P., TILL, U., HEPTINSTALL, S., KIRCHMAIER, C.M., SCHIRMER, A., MEYER, M. and BREDDIN, H.K. Actin polymerization during agonist-induced activation of human blood platelets. Blut, 60, 106, 1990. 23. SPANGENBERG, P., HEPTINSTALL, S., CRAWFORD, J. and TILL, U. Measurements of the G-/F-actin equilibrium in ADP-stimulated human platelets. Acta histochem., 4 1, 181-l 86, 1992. 24. VAN DEURS, B. and BEHNKE, 0. Membrane structure of nonactivated and activated human blood platelets as revealed by freeze-fracture: evidence for particle redistribution during platelet contraction. J. Cell Biol. 87, 209-218, 1980.
FUNCTIONAL PLATELET
STUDIES FUNCTION
ON PLATELETS OF A PATIENT WITH AN ACQUIRED DISORDER ASSOCIATED WITH AUTOANTIBODIES AGAINST MEMBRANE GLYCOPROTEIN IIB/IIIA COMPLEX
OF
P. Spangenberg’, C.M. Kirchmaier’, A. Schirmer’, M. Meyer3 and H.K. Breddin’ Institute of Pathological Biochemistry’ and Department of Genetics3 of the School of Medicine of Erfurt and Department of Angiology’ of the Center for Internal Medicine, J.W. Goethe University Frankfurt/M., F.R.G.
(Received 3.10.1992; accepted in revised form 23.11.1992 by Editor S. Lopaciuk) (Received by Executive Editorial Office 3.1.1993)
Abstract Platelet functions have been studied of a 63 year old woman with a severe acquired thrombopathy.The platelets did not adhere to siliconized glass. Aggregation could not be induced by either ADP (1 /IM) nor collagen (2 flglml), no release of serotonin was found under these conditions. Thrombin caused only a weak aggregation response. Quantitative analvsis of platelet actin revealed a very low total actin content (473 /./g/10’ platelets) .and an extremely low F-actin value 13% of total actin). Stimulation of platelets with 0.1 U/ml thrombin for 3 min resulted in an increase of only 5% ,F; actin, whereas ADP and collagen did not induce any actin polymerization. Ca movement in the patient’s platelets is severely impaired after ADP and collagen stimulation, whereas a normal Ca” movement was obtained by 0.1 U/ml thrombin. The inhibition of the functions of normal platelets (aggregation and actin polymerization) by addition of patient’s serum (5-10% final concentration) points to receptor blockade by platelet autoantibodies in the patient’s serum. The antibody was purified by adsorption on Protein-A-Sepharose. Addition of IgG-suspension (5% final concentration) to washed control platelets resulted in similar effects on aggregation and actin polymerization compared to the effects of patient’s serum.
Autoantibodies chronic against surface antigens are responsible for autoimmune thrombocytopenic purpura (l-3). The increased peripheral platelet destruction found in this disease is due to the cytotoxic effect of the autoantibodies toward the platelets. Frequently the antigenic determinants are associated with the glycoproteins Ilb/llla but the exact antigenic specificity is as Yet unknown. In autoimmune thrombocytopenic purpura also autoantibodies against glycoprotein lb were found (4). Until now they were underscored, but using a more sensitive method a recent study demonstrated the same frequency of glycoprotein lb autoantibodies as those against glycoproteins llblllla (5). The patient reported here shows a platelet function defect caused by autoantibodies towards the glycoprotein Ilb/llla complex without immunoinjury of the platelets (6). The patient exhibits a severe bleeding tendency which developed during life. We have investigated the functional behaviour of the patient’s platelets and the effects of patient’s serum and isolated IgG from Key words: platelets, autoantibodies, aggregation, calcium, actin, platelet dysfunction Corresponding author: P. Spangenberg, Institute of Pathological Biochemistry, School of Medicine of Erfurt, Nordhauser Str. 74, D-0-5010 Erfurt, F.R.G. 435
PLATELET AUTOANTIBODIES
patient’s
serum on the function
Vol. 69, No. 5
of platelets from healthy volunteers.
MATERIALS
AND METHODS
Blood was obtained from the patient and control subjects by venipuncture anticoagulated with 3.13% trisodium citrate (9 parts of blood and 1 part of trisodium Platelet rich plasma (PRP) was obtained by centrifugation platelets were prepared as previously described (7).
at 200
x Q for
15
min.
AQQreQatiOt'Iwas measured turbidometrically according to Born (8). Concentrations used in this study were 1 ,vM (ADP), 2 /./g/ml (collagen) and 0.1 U/ml (thrombin). The release reaction was measured by the amount of “C-5-hydroxytryptamine from the platelets that had been pre-labelled with this agent (9).
and was citrate). Washed
of agonists
(5HT)
released
Actin determinations were performed by the DNase I inhibition assay (10) in the modification Holme et al. (11) as previously described (12).
of
Calcium movements were followed measurements of INDO 1 /AM.
by
fluorometrically
according
to
Schirmer
et
al.
(7)
IgG from patient’s serum was isolated using a Protein-A-Sepharose column (Pharmacia, Sweden). The acetic acid-desorbed material was neutralized immediately and used in the replacement experiments.
RESULTS Case report. a 62 year-old female, was admitted to hospital in 1982 because of The patient Sch., paraproteinemia. In the same year unclear blood losses started and under the diagnoses of autoimmunohaemolytic anemia she was repeatedly treated with prednisolon and blood transfusions. In 1984 the haemorrhagic diathesis persisted with severe gastrointestinal bleeding In 1985 the diagnosis of idiopathic and epistaxis necessitating multiple blood transfusions. thrombocytopenic purpura was made and the patient was splenectomized. The low platelet count increased to subnormal and normal values, but the bleeding tendency persisted. In 1987 haemostatic tests indicated a normal platelet count, prolonged bleeding time, strongly reduced platelet adhesion and defective platelet aggregation. Aaareaation studies. Pig. 1 reports platelet aggregation responses obtained with patient’s PRP or washed platelets. Platelets did not aggregate with neither ADP (1 ,uM) nor collagen (2 pug/ml) and did not release considerable amounts of 5HT (4% release in ADP-induced platelets as compared to 16% release in control platelets using 1 PM ADP, 0% release in collagen-induced platelets as compared to 42% release in control platelets induced by 2 ,ug/ml collagen). Thrombin (0.1 U/ml) caused only a slight aggregation response compared to the thrombin-induced aggregation in control washed platelets. The platelets did not adhere to siliconized glass (index = 0.0, normal value 0.8 - 1.4). Patient’s serum induced a similar defect when mixed to control washed platelets. This effect was concentration-dependent and a complete inhibition by patient’s serum in case of ADP- or collagen-induced platelet aggregation was achieved at a concentration of 5% serum ((v/v), see Pig. 1). A similar effect was observed when patient’s serum was mixed to control washed This effect was also concentration-dependent, but only platelets and stimulated with thrombin. a partial inhibition resulted even in case of 5% patient’s serum ((v/v), see Pig. 1). A very similar behaviour was obtained using isolated IgG at a concentration of 5% (v/v), ADP- and collageninduced aggregation were completely blocked (data not shown here).
Vol. 69, No. 5
ADP
‘RP
PLATELET AUTOANTIBODIES
, collagen 7
I Irl------thrombin
HP
1
437
control
thrombin
WP
collagen
I
I
ADP
I
I
1:: patient
1
1
1
1
2 control
2
2 1
: patient’s
2
: control
serum serum
added added
FIG. 1. Aggregation of platelets from the patient and control subjects in the presence of patient’s serum. Patient’s platelets in plasma (PRP) were stimulated with ADP (1 PM) or collagen (2 /Ig/ml), washed patient’s platelets (WP) were stimulated with thrombin (0.1 U/ml). In the replacement experiments control washed platelets (WP) in the presence of control or patient’s serum (!5% final concentration) were subjected to thrombin (0.1 U/ml), collagen (2 pQ/ti) and ADP (I ,YM).
Calcium movements. As shown in Fio. 2 oatient’s platelets loaded with INDO l/AM and activated with thrombin or ionomycin exhibited ‘a quite normal increase of cytosolic Ca”, whereas those stimulated with ADP or collagen did not show any calcium signal. In the replacement experiments (control platelets in the presence of patient’s serum) this effect on the calcium movement was not obtained. ADP as well as collagen stimulated the typical Ca *’ increase in the cytoplasma as found in the case of control platelets in the presence of control serum (see Fig. 2, lower Part). Actin determinations. In the platelets of the patients extremely low value for F-actin normally towards stimulation lasted for 1 min, the cells were
a significantly smaller total actin content was measured and an was found (see Table I). These platelets did not polymerize actin with ADP, collagen or thrombin (stimulation with the agonists then lyzed and the actin was measured). ADP and collagen were
438
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
not able to induce any actin polymerization, whereas thrombin caused a polymerization of only 5% (the increase of F-actin found normally after thrombin-stimulation is about 30% of total actin).
+
tca2+1 - rca2fext - ca2f ext ext
+ tca2fext
rca2++ r&l
thrombin
1
300250-
200-
ADP
ionomycin
1
1
1
r-_
7,
‘!-T
0;
thrombin
collagen
5
0
T-7
1
0
1
min
collagen
_
nM
ADP
FIG. 2. Fluorescence tracings on activation of INDO 1 /AM-loaded platelets from the patient (upper part) and from a normal subject (lower part) in the absence (1) or presence (2) of patient’s serum (5% final concentration) with thrombin (0.1 U/ml), ADP (1 ,uM), collagen (2 pg/ml) or ionomycin (1 MM).
Actin polymerization was also studied in control washed platelets in the presence of patient’s serum (5% final concentration, v/v). As shown in Table II patient’s serum inhibits completely the whereas a weak actin polymerization is ADPand collagen-induced actin polymerization, detectable in thrombin-stimulated platelets. A similar behaviour is obtained when lgG (5% final concentration, v/v) is added to control platelets (data not shown here).
DISCUSSION As discussed in a previous paper (6) the platelet defect of the patient investigated here can be characterized as an acquired thrombopathy due to autoantibodies against the membrane glycoprotein Ilb/llla complex. The autoantibodies do not cause an increased peripheral platelet
Vol. 69, No. 5
PLATELET AUTOANTIBODIES
destruction, as it occurs in patients with chronic idiopathic thrombocytopenic shown by Meyer et al. (6) the autoantibodies are IgG-type antibodies.
TABLE Actin Analysis
total actin (pg 110’ platelets)
controls (n = 50)
patient
991
473
f
148
+ 12.3
97
F-actin (% of total actin)
20.2
+ 12.3
3
due to stimulation
ADP (1 PM) COLLAGEN (2 pglml) THROMBIN (0.1 U/ml)
As
of Platelets of the Patient and of Control Subjects
79.8
polymerization
(l-3).
I
G-actin (% of total actin)
Actin
purpura
of platelets
increase in F-actin (in % of total actin) controls (n = 5)
increase in F-actin (in % of total actin) patient
32 + 17 42 f 5 30-+ 1
0 0 5
Splenectomy of the patient who has shown originally an idiopathic thrombocytopenic purpura (6) induced a normalization of the platelet count (normal to subnormal platelet count), but haemorrhagic symptoms did not disappear. As described by Niessner et al. (13) and Balduini et al. (I 4) autoantibodies against the membrane glycoprotein llblllla complex with different complex with different antigenic determinants lead either to immunoinjury of platelets or to platelet function defects. Aggregation studies revealed a defective response to ADP, collagen and thrombin. Balduini et al. (14) reported a very similar case of a woman with autoantibodies against the glycoprotein Ilb/llla complex. Moreover, the patient’s serum induced the same defect when mixed with normal platelets, indicating that the autoantibody is present in serum and that the autoantibody is able to block the receptors of control platelets. The actin analysis of the patient’s platelets revealed atypical results. The total platelet actin content (473 ,eg/lO’ platelets) is less than half of the amount found in normal control platelets (991 + 148 pg/lO’ platelets). Whether this is the result of a defective actin equipment of platelets during thrombopoiesis or whether platelets possess a normal actin content which is not detectable due to the presence of actin-protein-complexes (actin-capping), which are still available in the platelet lysate and which may fail to inhibit the DNase I (biochemical assay of actin determination) remains open. The F-actin content of the platelets (3% of total platelet actin) is extremely low, since 20.2 + 12.1 % F-actin of total actin for females older than 35 years have been determined recently (12).
440
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
TABLE II Effect of Patient’s
Actin
Serum on Agonist-Induced
polymerization
due to stimulation
ADP (1 /JM) COLLAGEN (2 I.rglml) THROMBIN (0.1 U/ml)
Actin Polymerization
in Control
Platelets
of platelets
increase in F-actin (in % of total actin)
increase in F-actin (in % of total actin)
control platelets + control serum (n = 3)
control platelets + patient serum
18 * 27+ 30+
0 0 5
11 8 1
The F-actin content has been considered to be a parameter of platelet reactivity and donors with a high F-actin content (e.g. diabetics, about 40% F-actin) have been characterized to possess hyperreactive platelets (12). Consequently, the very low F-actin content of the patient investigated here is in correspondence to the defective aggregation behaviour. Physiological agonists rapidly induce a conversion of G- into F-actin (actin polymerization); e.g. thrombin increases within 15 set the F-actin content from 40 to 60-80 % (15-18). These experiments have been performed with washed platelets, which become preactivated during the preparation leading to increased basal F-actin values (12). Actin polymerization induced by ADP or collagen can also be studied in PRP and the increase in F-actin is agonist-dependent (see Table I, platelets of healthy volunteers). The autoantibodies against the Qlycoprotein Ilb/llla of the patient inhibit completely the ADP- and collagen-induced actin polymerization in the patient’s platelets, the thrombin response is very small (5% F-actin increase versus 30% in control platelets). Interestingly, the same effects occur when control platelets are mixed with patient’s serum. The ADP- and collagen-responses are completely inhibited again, whereas only 5% increase in F-actin was measurable when the cells were stimulated by thrombin. Since washed platelets were mixed with control serum and used for these experiments the increase in F-actin is smaller than the F-actin is smaller than the increase found in PRP during corresponding activation due to the prea.ctivation of platelets during the preparation (Table I versus Table II). The regulation of actin polymerization in platelets is still an open problem. The process seems to be receptor-mediated and the inhibitory effects of the llblllla autoantibodies on normal platelets described here support this statement. Moreover, transducer or effector systems might be involved in the regulation of actin polymerization (19, 20). We have shown recently that proteinkinase C and ,Ca” ions are somehow involved in the reorganization of microfjdaments in platelets and in the agonist-induced actin polymerization (21-23). Previous experiments Iwith phorbol ester and ionophore A 23187 failed to define a role for these intracellular mediators II 9, 20). The calcium movement experiments performed on the patient’s platelets would fit with the concept that calcium ions are involved in the regulation of actin polymerization since ADP- and collagen-induced increase of cytosolic Ca ‘+ is inhibited in those platelets (coincidental with the inhibition of aggregation and actin polymerization). The interpretation of the thrombin response however is more difficult since both aggregation and actin polymerization are inhibited but a normal calcium response is observed. Also the replacement experiments seem not to be congruent with the idea of the involvement of calcium ions in actin polymerization. The ADP-and collagen-induced calcium responses are quite normal in control platelets mixed with patient’s serum, although aggregation and actin polymerization are still completely inhibited. One possible
Vol. 69. No. 5
441
PLATELET AUTOANTIBODIES
explanation for this could be an internalization of the autoantibodies which may occur during the thrombopoiesis and which may inhibit the Ca” movement usually induced by collagen and ADP, whereas the autoantibodies in the replacement experiments do not become internalized (incubation times up to 1 h) and are not able to inhibit the Ca” movement. Although the F-actin content is very low in the platelets of the patient and actin polymerization is impaired the clot retraction is still normal (6). For clot retraction a structural coupling between actin, myosin and the fibrin network is essential. Van Deurs and Behnke (24) reported the existence of intramembrane particles which were evenly distributed in nonactivated platelets. The distribution of these particles did not change during platelet activation and aggregation (where actin polymerization occurs), but clusters of particles were found during clot retraction. It was suggested that this redistribution of the intramembrane particles reflects the mechanical transmembrane links between the actinlmyosin and the fibrin net. In conclusion, the Ilb/llla autoantibodies found in the patient described here cause dramatic defects in platelet function and mimic a similar situation when mixed with control platelets. These autoantibodies are therefore an interesting tool to study mechanisms of platelet activation.
Acknowledgements This work was supported by a grant from the International Institute of Thrombosis Diseases Frankfurt/M., F.R.G. and by a grant from the DFG No. llB5-Ki429/1-1
and Vascular
REFERENCES I. VAN LEEUWEN, E.F., VAN DER VEN, J.Th.M., ENGELFRIET, C.P. and VON DEM BORNE, A.E.G.Kr. Specificity of autoantibodies in autoimmune thrombocytopenenic purpura. Blood, 59, 23-26, 1982. 2. VARON, D. and KARPATKIN, S. A monoclonal anti-platelet antibody with decreased reactivity for autoimmune thrombocytopenic platelets. Proc. Natl. Acad. Sci. USA, 80, 69926998, 1983. 3. WOODS, V.L., OH, E.H., MASON, D. and MCMILLAN, R. Autoantibodies against the platelet glycoprotein Ilb/llla complex in patients with chronic ITP. Blood 63, 368-375, 1984. 4. WOODS, V.L., KURATA, Y., MONTGOMERY, R.R., TANI; P., MASON; D., OH, E.H. and MCMILLAN, R. Autoantibodies against platelet glycoprotein lb in patients with chronic immune thrombocytopenic purpura. Blood, 64, 156-l 60, 1984. 5. KIEFEL, V., SANTOSO, S., KAUFMANN, E. and MUELLER-ECKHARDT, C. Autoantibodies against glycoprotein lb/IX: a frequent finding in autoimmune thrombocytopenic purpura. Br. J. Haematol. 79, 256-262, 1991. 6. MEYER, M., KIRCHMAIER, CM., SCHIRMER, A., SPANGENBERG, P., STRGHL, Ch. and BREDDIN, H.K. Acquired thrombopathy: severe disorder of platelet function associated with autoantibodies against membrane glycoprotein Ilb-llla complex. 1. Glycoprotein analysis. Thrombos. Haemost., 65, 491-496, 1991. 7. SCHIRMER, A., BELLINGER, O.K. and KIRCHMAIER, C.M. Freies intrazytoplasmatisches Kalzium in Thrombozyten bei Gesunden und Patienten mit Thrombopathien. Inn. Med., 15, 151155, 1988. 8. BORN, G.V.R. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature, 194, 927-929, 1962. 9. HEPTINSTALL, S., BEAVEN, J., COCKBILL, S.R., HANLEY, S.P. and PARRY, M.J. Effects of a selective inhibitor of thromboxane svnthetase on human blood platelet behaviour. Thrombos. Res., 20, 219-230, 1980. 10. BLIKSTAD, I., MARKEY, F., CARLSSON, L., PERSSON, S. and LINDBERG, U. Selective assay of monomeric and filamentous actin in cell extracts. Cell, 15, 935-943, 1978. 11. HOLME, T., KELLIE, S., WYKE, J.A. and CRAWFORD, N. Effect of transforamtion by Rous sarcoma virus on the character and distribution of actin in rat-l fibroblasts: a biochemical and microscopical study. 8r. J. Cancer, 53, 465-476, 1986.
442
PLATELET AUTOANTIBODIES
Vol. 69, No. 5
12. SPANGENBERG, P. and TILL, U. Actin filament content in platelets - a sensitive index of cellular reactivity. Biosc. Rep., 9, 307-313, 1989. 13. NIESSNER, H., CLEMETSON, K.J., PANZER, S., MUELLER-ECKHARDT, Ch., SANTOSO, S. and BETTELHEIM, P. Acquired thrombasthenia due to GP Ilb/llla-specific autoantibodies. Blood, 68, 571-576, 1986. 14. BALDUINI, C.L., GRIGNANI, G., SINIGAGLIA, F., BISIO, A., PACCHIARINI, L., ROTA SCALABRINI, D., BALDUINI, C., MAURI, C. and ASCARI, E. Severe platelet dysfunction in a patient with autoantibodies against membrane glycoprotein Ilb/llla. Haemost., 7, 98-l 04, 1987. 15. CARLSSON, L., MARKEY, F., BLIKSTAD, I., PERSSON, T. and LINDBERG; U. Reorganization of actin in platelets stimulated by thrombin as measured by the DNase I inhibition assay. Proc. Natl. Acad. Sci. USA, 76, 6376-6380, 1979. 16. FOX, E.J.B. and PHILLIPS, D.R., Inhibition of actin polymerization in blood platelets by cytochalasins. Nature, 292, 650-652, 1981. 17. CASELLA, J.F., FLANAGAN, M.D. and LIN, S. Cytochalasin D inhibits actin polymerization and induces depolymerization of actin filaments formed during platelet shape change. Nature, 293, 302-305, 1981. 18. PRIBLUDA, V. and ROTMAN, A. Dynamics of membrane-cytoskeleton interaction in activated blood platelets. Biochem., 2 1, 2825-2832, 1982. 19. LASSING, I. and LINDBERG, U. Specific interaction between phosphatidylinositol-4,5bisphosphate and profilactin. Nature, 314, 473-474, 1985. 20. LASSING, I. and LINDBERG, U. Evidence that the phosphatidylinositol cycle is linked to cell motility. Exp. Cell Res., 174, l-l 5, 1988. 21. SPANGENBERG, P., KOTELIANSKY, V.E., LUKASHEV, M.E. and TILL, U. EinfluB von Phorbolester (PMA) und Forskolin auf das G-/F-Aktin Gleichgewicht humaner Thrombozyten. Acta histochem., 39, 397-402, 1990. 22. SPANGENBERG, P., TILL, U., HEPTINSTALL, S., KIRCHMAIER, C.M., SCHIRMER, A., MEYER, M. and BREDDIN, H.K. Actin polymerization during agonist-induced activation of human blood platelets. Blut, 60, 106, 1990. 23. SPANGENBERG, P., HEPTINSTALL, S., CRAWFORD, J. and TILL, U. Measurements of the G-/F-actin equilibrium in ADP-stimulated human platelets. Acta histochem., 4 1, 181-l 86, 1992. 24. VAN DEURS, B. and BEHNKE, 0. Membrane structure of nonactivated and activated human blood platelets as revealed by freeze-fracture: evidence for particle redistribution during platelet contraction. J. Cell Biol. 87, 209-218, 1980.