Immunology 1976 31 595
C1 and human platelets III.
ROLE OF C1 SUBCOMPONENTS IN PLATELET AGGREGATION INDUCED BY AGGREGATED IgG
J. L. WAUTIER, H. SOUCHON, L. COHEN SOLAL, A. P. PELTIER & J. P. CAEN ERA 335 CNRS, INSERM U150 et U18, H6pital Lariboisiere, rue Ambroise Pare', Paris, Cedex 10, France
Received 16 February 1976; accepted for publication
1
April 1976
this study we have tried to investigate the relationships between the C1 subcomponents and platelets and to ascertain their possible role in platelet aggregation induced by aggregated IgG.
Summary. Studies have been performed with platelets using C1 haemolytic assays and platelet aggregation induced by anti-Clq, anti-Cls and aggregated IgG in the presence of C1 subcomponents Clq, Clr and Cis. Clq was removed by EDTA or modified by collagenase from human platelets while after the same treatment Cls remained bound to the platelets. EDTA treated platelets were no longer aggregated by aggregated IgG. The addition of Clq restored the reactivity of the platelets to aggregated IgG while the addition of Clr or Cls was without effect. Furthermore, the addition of CIr or Cls to Clq inhibited the action of Clq in platelet aggregation induced by IgG. The possible association between the different C1 subcomponents and human platelets is discussed.
MATERIALS AND METHODS C1 subcomponents (a) C1 was prepared as previously described (Wautier et al., 1974). (b) In preliminary experiments C1 subcomponents of the serum (Clq, Cir, Cls) were separated and tested in haemolytic assay according to the technique of Lepow, Naff, Todd, Pensky & Hinz (1963). (c) Pure preparation of Clq, Cir and Cls were obtained by affinity chromatography of whole human serum on human IgG coupled to Sepharose 4B followed by preparative electrophoresis for Cls, by chromatography on DEAE-cellulose for Cir, and by chromatography on DEAE-cellulose and gel filtration on Sephadex G-200 for Clq (Assimeth, Bing & Painter, 1974). (d) Purified Clq was also obtained by successive chromatography on DEAE-cellulose, Sephadex G-200, and CM cellulose according to the method described by Reid, Lowe & Porter (1972) as later
INTRODUCTION In previous communications (Wautier, Tobelem, Peltier & Caen, 1974, Wautier, Tobelem, Peltier & Caen, 1976) we have demonstrated an association of C1 to human platelets and described a role for C1 in platelet aggregation induced by aggregated IgG. In Correspondence: Dr J. L. Wautier, ERA 335 CNRS, INSERM U150 et U18, H6pital Lariboisiere, 2 rue Ambroise Pare, 75475 Paris, Cedex 10, France.
595
596
J. L. Wautier et al.
modified by Reid (1974). The purified Clq was free of contaminants. The different components were concentrated by AMICON ultrafiltration and tested for purity in Ouchterlony analysis, immunoelectrophoresis and analytical disc electrophoresis. The protein concentration was adjusted to 1 mg/ ml for Clq, 50 ,ug/ml for CIr and 1X5 mg/ml for Cls. (e) Incubation of Clq with collagenase: bacterial collagenase (grade B, Calbiochem), contained a major component and two minor components as shown by polyacrylamide disc electrophoresis. It had no esterolytic activity on either BAEE (benzoylarginine ethyl ester) or ATEE (acetyltyrosine ethyl ester) and its specific activity was 101 u/g. Clq (1 mg) was incubated for 16 h at 370 with 100 ufml of the collagenase at pH 7-5 in Tris-HCl buffer 0-015 M containing CaCI2 (001 M). At the end of the digestion insoluble purified collagen was added to the mixture and incubated for 4 h at 37°, pH 5 4, to remove excess collagenase. The collagen was eliminated by filtration on a Millipore filter and the solution dialysed against Tris-HCl buffer, 0-015 M, pH 7-4 at 40. (f) Incubation of C1 with collagenase: Cl, at different concentrations, was incubated with collagenase (200 u/ml) and then tested for its haemolytic activity and for its esterolytic activity on ATEE. The latter determination was performed using a titrigraph Radiometer (Copenhagen, Denmark) using the procedure of Legrand, Caen, Booyse, Rafelson & Robert (1973).
continuous stirring at 100 r.p.m. using a Brystonaggregometer (Labintec, Montpellier, France). Ten or 40 4u1 of anti-Clq, anti-Cls (Behringwerke Marburg) or aggregated IgG (100-150 ,g), prepared as previously described by Wautier et a. (1976), were added to the gel-filtered platelets. The antisera anti-Clq, anti-Cls were monospecific and platelet aggregation induced by these antisera was not inhibited by monomeric IgG. Platelet preparations: gel-filtered platelets, or EDTA, or collagenase treated GFP were also tested in platelet aggregation. C1 subcomponents used for platelet aggregation: native Clq (40-85 pg), or collagenase treated Clq (85 pg), CIr (1 ug), Cis (60 pg) were added to these platelet preparations in the presence of aggregated IgG or anti-Clq anti-Cls antisera. Platelet aggregation was expressed in terms of velocity and intensity; velocity was measured by drawing a tangent to the steepest part of the aggregation curve and measuring the increase in transmittance that occurs after 30 s. The intensity is the height of the curve measured at 2 min. Both the velocity and intensity are expressed as the percentage of the maximum transmission given by the buffer itself (Caen, Larrieu & Samama, 1975).
RESULTS Platelet C1 subcomponent in haemolytic assay
Platelets
(a) Platelet solution. Platelets were isolated by gel filtration on Sepharose 2B (Wautier et al., 1974).
(b) Dissociation ofplatelet associated C1. Platelets treated by EDTA (Wautier et al., 1976) and then tested in C1 haemolytic assay in the presence of each of the different C1 subcomponents (Clq, Cir, Cis) alone or reassociated according to the technique described by Lepow et al. (1963). The results are expressed as the percentage of the haemolytic activity of the Clq+Clr+Cls mixture. Treatment by collagenase: GFP were incubated with collagenase (200 u/m final concentration) for 5 min at 220 and then tested in platelet aggregation. were
(c) Platelet aggregation. Platelet aggregation was studied by a photometric method at 370 and with
EDTA-treated platelets were tested for their C1 haemolytic activity after the addition of the one or two of the C1 subcomponents. While each of the serum C1 subcomponents alone or associated in pairs did not possess any haemolytic activity, a C1 haemolytic activity was present when the three subcomponents were reassociated. The results observed with the platelet pellet obtained following the EDTA treatment of GFP, and which was resuspended in a Ca2 +- and Mg2 +-containing medium showed that the only association which possessed a major and constant C1 haemolytic activity was the platelet pellet + Clq + Clr which means that the platelet pellet still contains Cls. The results of four experiments are summarized in Table 1. The results are expressed as the percentage of the C1 hemolytic activity obtained, the reassociated Clq+Clr+Cls being considered as 100 per cent.
Cl and human platelets. III Table 1. Cl haemolytic assays of EDTA-treated platelets re-associated with the different Cl-subcomponents Expt. number
1
2
3
4
0
0
0
0
80
100
80
75
+Clq+Cls
0
10
20
0
Platelet pellet +Clr+Cls Clq+Clr+Cls
20
0
0
0
Platelet pellet Platelet pellet
+Clq+Clr Platelet pellet
100
Results are expressed as the percentage of the haemolytic activity of the Clq+Clr+ CIs mixture. Cl subcomponents were prepared according to the technique of (Lepow 1963).
597
not occur. When COr or Cls were added to the mixture Clq + EDTA treated platelets, the aggregation induced by anti-Clq was diminished.
Incubation of C1 with collagenase. C1 incubated with collagenase lost its haemolytic activity but the esterolytic activity as tested with ATEE was not modified.
Induced by aggregated IgG (Table 3) GFP incubated with EDTA and resuspended in a Ca2 +- and Mg2 +-containing medium, were no longer reactive with aggregated IgG. The aggregation was restored by the addition of native Clq while collagenase treated Clq was less effective and Cir and Cls were without effect. The addition of Cir to native Clq diminished its activity which was completely abolished by the addition of CIs or Cir + CIs.
Platelet aggregation
Induced by anti-Cls (Table 2) GFP were well aggregated by anti-Cls and remained capable of reacting with anti-Cls after EDTA or collagenase treatment. This reaction was partially inhibited by the addition of Clq and COr and almost completely by the addition of Cls. Table 2. Platelet aggregation induced by anti-Cls (10 Al) (mean of five experiments)
Preparations GFP control EDTA-treated GFP Collagenase-treated GFP EDTA-treated GFP +Clq +Clr +Cls
Lag phase Velocity Intensity (s) (per cent) (per cent) 30 40
39 24
85 60
30
25
68
30 50 40
16 24 11
34 50 5
Induced by anti-Clq As we have previously described, control GFP were aggregated by anti Clq. This aggregation was abolished by previous treatment of the platelets by EDTA or collagenase. The addition of Clq to the treated platelets restored their capacity to react with antiClq while COr or Cls were without effect. When the Clq had been previously incubated with collagenase, platelet aggregation in the presence of anti Ctq did
Table 3. Platelet aggregation induced by aggregated IgG (100 jg) in the presence of Cl subcomponents (mean of six experiments) Lag phase Velocity Intensity (s) (per cent) (percent)
Preparations Buffer EDTA-treated GFP: +Clq + eollagenase-treated Clq + Clr +Cls +Clq4-Clr +Clq+Cls +Clr+Cls +Clq FClr+Cls
c00
0
0
100
20
75
252
6 0 0 6 0 0 0
11 0 0 18 0 0 0
0
o
280 o0 ox ox
DISCUSSION It has been firmly established that Cl, the first component of the complement system, is made up of three subcomponents Clq, Cir, Cls which are held together by calcium (Lepow et al., 1963). A fourth subcomponent Clt has been recently identified but its role is not completely elucidated (Assimeth & Painter, 1975). C1 fixation to the Fc fragment of the antibody moiety of an antigen-antibody complex proceeds through Clq (Muiller-Eberhard & Kunkel, 1961) and perhaps Cir (Assimeth & Painter, 1975),
598
J. L. Wautier et al.
this attachment results in an indirect activation of Clr and Cls. The results obtained in this work indicate that platelet-associated CI is bound to the platelet through its Cls subcomponent and that Clq and COr are bound to Cls and are not themselves directly attached to the platelet. Direct attachment of Cls to the platelet is demonstrated by the fact that EDTA-treated platelets no longer have C1 haemolytic activity which is restored by the addition of Clq, CIr and calcium to the EDTA-treated platelets. The same EDTA-treated platelets were still aggregated by anti-Cls but not by anti-Clq and the anti-Clq induced aggregation was restored by the addition of Clq. The presence of Clq was necessary for the aggregation of platelets by aggregated IgG. When Clq was pre-incubated with collagenase and mixed with EDTA-treated platelets, it was no longer effective in platelet aggregation induced by anti-Clq and aggregated IgG. This suggests that Clq is attached to the plateletbound Cls by its collagen-like moiety, which would leave its non-collagen portion free to react with aggregated IgG. Knobel, Heusser, Rodrick & Isliker (1974) have similarly shown that Clq incubated with collagenase loses its C1 haemolytic activity but maintains its capacity of reacting with aggregated IgG. Collagenase contained in some platelet granules (Chesney, Harper & Coman, 1974) may inhibit platelet-associated Clq as bacterial collagenase but the two enzymes are different in their proteolytic activity and the presence of anti-protease in platelet and in plasma would counteract this activity. While the COr subcomponent is necessary for platelet C1 haemolytic activity, it does not seem to be required in aggregated IgG induced platelet aggregation since this aggregation may occur in the absence of any detectable COr. The use of an antiCOr serum may be necessary to confirm this interpretation. Finally other results indicate that an optimal platelet C1 haemolytic activity or aggregation induced by aggregated IgG requires a critical Clq/Clr/ CIs ratio. In this sense the mixtures Clq + Cir + Cis or Clq+Cls or Clq+Clr were less effective than Clq alone in this type of platelet aggregation. This may be explained by a competition between Cls still present on platelets and Cis added to the mixture for the part of the molecule of Clq that reacts with Cls.
It therefore appears that Cls is directly bound to platelets but no data are available concerning its binding site. Cazenave et al. (1976) have observed that Clq inhibited platelet aggregation induced by collagen and postulate that Clq competes with collagen for a specific site on platelets. These results are in accordance with the fact that the collagen-like moiety of Clq reacts with platelets and Cls. A possible interaction between collagen and platelet-associated Cls is under investigation. Speculations can be made about the significance of platelet-associated C1. The activation of Cls associated to platelets can be produced indirectly by the interaction of Clq with IgG and this activation may be a starting event in platelet aggregation induced by immune complexes. The enzymatic activity of Cls may act on some platelet-aggregating systems. The inhibitory activity of the synthetic substrate (ATEE) of Cls on platelet aggregation may be a further indication for a role of Cls in platelet aggregation.
ACKNOWLEDGMENT
This work was supported by a grant of INSERM ATP 12/74/33. REFERENCES ASSIMETH S.N., BING D.H. & PAINTER R.H. (1974) A simple method for isolation of the first component of complement by affinity chromatography. J. Immunol. 113, 225. AssiMETH S.N. & PAINTER R.H. (1975) The macromolecular structure of the first component of complement. J. Immunol. 115, 488. CAEN J., LARRIEU M.-J. & SAMAMA M. (1975) L'himostaseMethodes d'Exploration et Diagnostic pratique, 2nd edn. Expansion Scientifique Franqaise, Paris. CAZENAVE J.-P., AssiMETH S.N., PAINTER R.H., PACKHAM M.A., MUSTARD J.F. (1976) Clq inhibition of the interaction of collagen with human platelets. J. Immunol. 116, 162. CHESNEY C., HARPER E. & COLMAN R.W. (1974) Human platelet collagenase. J. Clin. Invest. 53, 1647. KNOBEL H.R., HEUSSER C.H., RODRICK M.L. & ISLIKER H. (1974) Enzymatic digestion of the first component of human complement (Clq). J. Immunol. 112, 2094. LEGRAND Y., CAEN J., BOOYSE F.M., RAFELSON M.E. & ROBERT B. (1973) Studies on a human blood platelet protease with elastolytic activity. Biochim. biophys. Acta (Amst.), 309, 406.
Cl and human platelets. III LEPOW J.H., NAFF G.B., TODD E.W., PENSKY J. & HINZ C.F. (1963) Chromatographic resolution of the first component of human complement into three activities. J. exp. Med.
117, 983. MULLER-EBERHARD H.J. & KUNKEL H.G. (1961) Isolation of a thermolabile serum protein which precipitates with gamma-globulin aggregates and participates in immunohemolysis. Proc. Soc. exp. Biol. (N. Y.), 106, 291. REID K.B.M., LOWE D.M. & PORTER R.R. (1972) Isolation and characterization of Clq subcomponent of the first component of complement from human and rabbit sera.
599
Biochemistry, 130, 749. REID K.B.M. (1974) A collagen like amino acid sequence in a polypeptide chain of human Clq. Biochemistry, 141, 189. WAUTIER J.-L., TOBELEM G.M., PELTIER A.P. & CAEN J.P. (1974) Evidence for Cl on human platelets. Haemostasis, 2, 281. WAUTIER J.-L., TOBELEM G.M., PELTIER A.P. & CAEN J.P. (1976) C1 and human platelets. Detection by immunological methods and role. Immunology, 30, 459.
C1 and human platelets III.
ROLE OF C1 SUBCOMPONENTS IN PLATELET AGGREGATION INDUCED BY AGGREGATED IgG
J. L. WAUTIER, H. SOUCHON, L. COHEN SOLAL, A. P. PELTIER & J. P. CAEN ERA 335 CNRS, INSERM U150 et U18, H6pital Lariboisiere, rue Ambroise Pare', Paris, Cedex 10, France
Received 16 February 1976; accepted for publication
1
April 1976
this study we have tried to investigate the relationships between the C1 subcomponents and platelets and to ascertain their possible role in platelet aggregation induced by aggregated IgG.
Summary. Studies have been performed with platelets using C1 haemolytic assays and platelet aggregation induced by anti-Clq, anti-Cls and aggregated IgG in the presence of C1 subcomponents Clq, Clr and Cis. Clq was removed by EDTA or modified by collagenase from human platelets while after the same treatment Cls remained bound to the platelets. EDTA treated platelets were no longer aggregated by aggregated IgG. The addition of Clq restored the reactivity of the platelets to aggregated IgG while the addition of Clr or Cls was without effect. Furthermore, the addition of CIr or Cls to Clq inhibited the action of Clq in platelet aggregation induced by IgG. The possible association between the different C1 subcomponents and human platelets is discussed.
MATERIALS AND METHODS C1 subcomponents (a) C1 was prepared as previously described (Wautier et al., 1974). (b) In preliminary experiments C1 subcomponents of the serum (Clq, Cir, Cls) were separated and tested in haemolytic assay according to the technique of Lepow, Naff, Todd, Pensky & Hinz (1963). (c) Pure preparation of Clq, Cir and Cls were obtained by affinity chromatography of whole human serum on human IgG coupled to Sepharose 4B followed by preparative electrophoresis for Cls, by chromatography on DEAE-cellulose for Cir, and by chromatography on DEAE-cellulose and gel filtration on Sephadex G-200 for Clq (Assimeth, Bing & Painter, 1974). (d) Purified Clq was also obtained by successive chromatography on DEAE-cellulose, Sephadex G-200, and CM cellulose according to the method described by Reid, Lowe & Porter (1972) as later
INTRODUCTION In previous communications (Wautier, Tobelem, Peltier & Caen, 1974, Wautier, Tobelem, Peltier & Caen, 1976) we have demonstrated an association of C1 to human platelets and described a role for C1 in platelet aggregation induced by aggregated IgG. In Correspondence: Dr J. L. Wautier, ERA 335 CNRS, INSERM U150 et U18, H6pital Lariboisiere, 2 rue Ambroise Pare, 75475 Paris, Cedex 10, France.
595
596
J. L. Wautier et al.
modified by Reid (1974). The purified Clq was free of contaminants. The different components were concentrated by AMICON ultrafiltration and tested for purity in Ouchterlony analysis, immunoelectrophoresis and analytical disc electrophoresis. The protein concentration was adjusted to 1 mg/ ml for Clq, 50 ,ug/ml for CIr and 1X5 mg/ml for Cls. (e) Incubation of Clq with collagenase: bacterial collagenase (grade B, Calbiochem), contained a major component and two minor components as shown by polyacrylamide disc electrophoresis. It had no esterolytic activity on either BAEE (benzoylarginine ethyl ester) or ATEE (acetyltyrosine ethyl ester) and its specific activity was 101 u/g. Clq (1 mg) was incubated for 16 h at 370 with 100 ufml of the collagenase at pH 7-5 in Tris-HCl buffer 0-015 M containing CaCI2 (001 M). At the end of the digestion insoluble purified collagen was added to the mixture and incubated for 4 h at 37°, pH 5 4, to remove excess collagenase. The collagen was eliminated by filtration on a Millipore filter and the solution dialysed against Tris-HCl buffer, 0-015 M, pH 7-4 at 40. (f) Incubation of C1 with collagenase: Cl, at different concentrations, was incubated with collagenase (200 u/ml) and then tested for its haemolytic activity and for its esterolytic activity on ATEE. The latter determination was performed using a titrigraph Radiometer (Copenhagen, Denmark) using the procedure of Legrand, Caen, Booyse, Rafelson & Robert (1973).
continuous stirring at 100 r.p.m. using a Brystonaggregometer (Labintec, Montpellier, France). Ten or 40 4u1 of anti-Clq, anti-Cls (Behringwerke Marburg) or aggregated IgG (100-150 ,g), prepared as previously described by Wautier et a. (1976), were added to the gel-filtered platelets. The antisera anti-Clq, anti-Cls were monospecific and platelet aggregation induced by these antisera was not inhibited by monomeric IgG. Platelet preparations: gel-filtered platelets, or EDTA, or collagenase treated GFP were also tested in platelet aggregation. C1 subcomponents used for platelet aggregation: native Clq (40-85 pg), or collagenase treated Clq (85 pg), CIr (1 ug), Cis (60 pg) were added to these platelet preparations in the presence of aggregated IgG or anti-Clq anti-Cls antisera. Platelet aggregation was expressed in terms of velocity and intensity; velocity was measured by drawing a tangent to the steepest part of the aggregation curve and measuring the increase in transmittance that occurs after 30 s. The intensity is the height of the curve measured at 2 min. Both the velocity and intensity are expressed as the percentage of the maximum transmission given by the buffer itself (Caen, Larrieu & Samama, 1975).
RESULTS Platelet C1 subcomponent in haemolytic assay
Platelets
(a) Platelet solution. Platelets were isolated by gel filtration on Sepharose 2B (Wautier et al., 1974).
(b) Dissociation ofplatelet associated C1. Platelets treated by EDTA (Wautier et al., 1976) and then tested in C1 haemolytic assay in the presence of each of the different C1 subcomponents (Clq, Cir, Cis) alone or reassociated according to the technique described by Lepow et al. (1963). The results are expressed as the percentage of the haemolytic activity of the Clq+Clr+Cls mixture. Treatment by collagenase: GFP were incubated with collagenase (200 u/m final concentration) for 5 min at 220 and then tested in platelet aggregation. were
(c) Platelet aggregation. Platelet aggregation was studied by a photometric method at 370 and with
EDTA-treated platelets were tested for their C1 haemolytic activity after the addition of the one or two of the C1 subcomponents. While each of the serum C1 subcomponents alone or associated in pairs did not possess any haemolytic activity, a C1 haemolytic activity was present when the three subcomponents were reassociated. The results observed with the platelet pellet obtained following the EDTA treatment of GFP, and which was resuspended in a Ca2 +- and Mg2 +-containing medium showed that the only association which possessed a major and constant C1 haemolytic activity was the platelet pellet + Clq + Clr which means that the platelet pellet still contains Cls. The results of four experiments are summarized in Table 1. The results are expressed as the percentage of the C1 hemolytic activity obtained, the reassociated Clq+Clr+Cls being considered as 100 per cent.
Cl and human platelets. III Table 1. Cl haemolytic assays of EDTA-treated platelets re-associated with the different Cl-subcomponents Expt. number
1
2
3
4
0
0
0
0
80
100
80
75
+Clq+Cls
0
10
20
0
Platelet pellet +Clr+Cls Clq+Clr+Cls
20
0
0
0
Platelet pellet Platelet pellet
+Clq+Clr Platelet pellet
100
Results are expressed as the percentage of the haemolytic activity of the Clq+Clr+ CIs mixture. Cl subcomponents were prepared according to the technique of (Lepow 1963).
597
not occur. When COr or Cls were added to the mixture Clq + EDTA treated platelets, the aggregation induced by anti-Clq was diminished.
Incubation of C1 with collagenase. C1 incubated with collagenase lost its haemolytic activity but the esterolytic activity as tested with ATEE was not modified.
Induced by aggregated IgG (Table 3) GFP incubated with EDTA and resuspended in a Ca2 +- and Mg2 +-containing medium, were no longer reactive with aggregated IgG. The aggregation was restored by the addition of native Clq while collagenase treated Clq was less effective and Cir and Cls were without effect. The addition of Cir to native Clq diminished its activity which was completely abolished by the addition of CIs or Cir + CIs.
Platelet aggregation
Induced by anti-Cls (Table 2) GFP were well aggregated by anti-Cls and remained capable of reacting with anti-Cls after EDTA or collagenase treatment. This reaction was partially inhibited by the addition of Clq and COr and almost completely by the addition of Cls. Table 2. Platelet aggregation induced by anti-Cls (10 Al) (mean of five experiments)
Preparations GFP control EDTA-treated GFP Collagenase-treated GFP EDTA-treated GFP +Clq +Clr +Cls
Lag phase Velocity Intensity (s) (per cent) (per cent) 30 40
39 24
85 60
30
25
68
30 50 40
16 24 11
34 50 5
Induced by anti-Clq As we have previously described, control GFP were aggregated by anti Clq. This aggregation was abolished by previous treatment of the platelets by EDTA or collagenase. The addition of Clq to the treated platelets restored their capacity to react with antiClq while COr or Cls were without effect. When the Clq had been previously incubated with collagenase, platelet aggregation in the presence of anti Ctq did
Table 3. Platelet aggregation induced by aggregated IgG (100 jg) in the presence of Cl subcomponents (mean of six experiments) Lag phase Velocity Intensity (s) (per cent) (percent)
Preparations Buffer EDTA-treated GFP: +Clq + eollagenase-treated Clq + Clr +Cls +Clq4-Clr +Clq+Cls +Clr+Cls +Clq FClr+Cls
c00
0
0
100
20
75
252
6 0 0 6 0 0 0
11 0 0 18 0 0 0
0
o
280 o0 ox ox
DISCUSSION It has been firmly established that Cl, the first component of the complement system, is made up of three subcomponents Clq, Cir, Cls which are held together by calcium (Lepow et al., 1963). A fourth subcomponent Clt has been recently identified but its role is not completely elucidated (Assimeth & Painter, 1975). C1 fixation to the Fc fragment of the antibody moiety of an antigen-antibody complex proceeds through Clq (Muiller-Eberhard & Kunkel, 1961) and perhaps Cir (Assimeth & Painter, 1975),
598
J. L. Wautier et al.
this attachment results in an indirect activation of Clr and Cls. The results obtained in this work indicate that platelet-associated CI is bound to the platelet through its Cls subcomponent and that Clq and COr are bound to Cls and are not themselves directly attached to the platelet. Direct attachment of Cls to the platelet is demonstrated by the fact that EDTA-treated platelets no longer have C1 haemolytic activity which is restored by the addition of Clq, CIr and calcium to the EDTA-treated platelets. The same EDTA-treated platelets were still aggregated by anti-Cls but not by anti-Clq and the anti-Clq induced aggregation was restored by the addition of Clq. The presence of Clq was necessary for the aggregation of platelets by aggregated IgG. When Clq was pre-incubated with collagenase and mixed with EDTA-treated platelets, it was no longer effective in platelet aggregation induced by anti-Clq and aggregated IgG. This suggests that Clq is attached to the plateletbound Cls by its collagen-like moiety, which would leave its non-collagen portion free to react with aggregated IgG. Knobel, Heusser, Rodrick & Isliker (1974) have similarly shown that Clq incubated with collagenase loses its C1 haemolytic activity but maintains its capacity of reacting with aggregated IgG. Collagenase contained in some platelet granules (Chesney, Harper & Coman, 1974) may inhibit platelet-associated Clq as bacterial collagenase but the two enzymes are different in their proteolytic activity and the presence of anti-protease in platelet and in plasma would counteract this activity. While the COr subcomponent is necessary for platelet C1 haemolytic activity, it does not seem to be required in aggregated IgG induced platelet aggregation since this aggregation may occur in the absence of any detectable COr. The use of an antiCOr serum may be necessary to confirm this interpretation. Finally other results indicate that an optimal platelet C1 haemolytic activity or aggregation induced by aggregated IgG requires a critical Clq/Clr/ CIs ratio. In this sense the mixtures Clq + Cir + Cis or Clq+Cls or Clq+Clr were less effective than Clq alone in this type of platelet aggregation. This may be explained by a competition between Cls still present on platelets and Cis added to the mixture for the part of the molecule of Clq that reacts with Cls.
It therefore appears that Cls is directly bound to platelets but no data are available concerning its binding site. Cazenave et al. (1976) have observed that Clq inhibited platelet aggregation induced by collagen and postulate that Clq competes with collagen for a specific site on platelets. These results are in accordance with the fact that the collagen-like moiety of Clq reacts with platelets and Cls. A possible interaction between collagen and platelet-associated Cls is under investigation. Speculations can be made about the significance of platelet-associated C1. The activation of Cls associated to platelets can be produced indirectly by the interaction of Clq with IgG and this activation may be a starting event in platelet aggregation induced by immune complexes. The enzymatic activity of Cls may act on some platelet-aggregating systems. The inhibitory activity of the synthetic substrate (ATEE) of Cls on platelet aggregation may be a further indication for a role of Cls in platelet aggregation.
ACKNOWLEDGMENT
This work was supported by a grant of INSERM ATP 12/74/33. REFERENCES ASSIMETH S.N., BING D.H. & PAINTER R.H. (1974) A simple method for isolation of the first component of complement by affinity chromatography. J. Immunol. 113, 225. AssiMETH S.N. & PAINTER R.H. (1975) The macromolecular structure of the first component of complement. J. Immunol. 115, 488. CAEN J., LARRIEU M.-J. & SAMAMA M. (1975) L'himostaseMethodes d'Exploration et Diagnostic pratique, 2nd edn. Expansion Scientifique Franqaise, Paris. CAZENAVE J.-P., AssiMETH S.N., PAINTER R.H., PACKHAM M.A., MUSTARD J.F. (1976) Clq inhibition of the interaction of collagen with human platelets. J. Immunol. 116, 162. CHESNEY C., HARPER E. & COLMAN R.W. (1974) Human platelet collagenase. J. Clin. Invest. 53, 1647. KNOBEL H.R., HEUSSER C.H., RODRICK M.L. & ISLIKER H. (1974) Enzymatic digestion of the first component of human complement (Clq). J. Immunol. 112, 2094. LEGRAND Y., CAEN J., BOOYSE F.M., RAFELSON M.E. & ROBERT B. (1973) Studies on a human blood platelet protease with elastolytic activity. Biochim. biophys. Acta (Amst.), 309, 406.
Cl and human platelets. III LEPOW J.H., NAFF G.B., TODD E.W., PENSKY J. & HINZ C.F. (1963) Chromatographic resolution of the first component of human complement into three activities. J. exp. Med.
117, 983. MULLER-EBERHARD H.J. & KUNKEL H.G. (1961) Isolation of a thermolabile serum protein which precipitates with gamma-globulin aggregates and participates in immunohemolysis. Proc. Soc. exp. Biol. (N. Y.), 106, 291. REID K.B.M., LOWE D.M. & PORTER R.R. (1972) Isolation and characterization of Clq subcomponent of the first component of complement from human and rabbit sera.
599
Biochemistry, 130, 749. REID K.B.M. (1974) A collagen like amino acid sequence in a polypeptide chain of human Clq. Biochemistry, 141, 189. WAUTIER J.-L., TOBELEM G.M., PELTIER A.P. & CAEN J.P. (1974) Evidence for Cl on human platelets. Haemostasis, 2, 281. WAUTIER J.-L., TOBELEM G.M., PELTIER A.P. & CAEN J.P. (1976) C1 and human platelets. Detection by immunological methods and role. Immunology, 30, 459.
