- Email: [email protected]
Platelet-Collagen Adhesion — Membrane Fluidity and the Development of High Affinity Adhesion through Multiple Interacting Sites
Collagen Re!. Res. Vo!. 111981, p. 517-526
Platelet-Collagen Adhesion - Membrane Fluidity and the Development of High Affinity Adhesion through Multiple Interacting Sites SAMUEL A. SANTOROl and LEON
w.
CUNNINGHAM2
Division of Laboratory Medicine, Washington University Schoo! of Medicine, St. Louis, MO 63110, USA 2 Department of Biochemistry, Vanderbilt University Schoo! of Medicine, Nashville, TN 37232, USA 1
Abstract Multiple, linked interactions between the platelet surface and collagen fibers have been implicated in the initiation of platelet secretion and subsequent aggregation. The formation of such multiple simultaneous interactions could give rise to high affinity adhesion of platelets to collagen even though the affinity of thc individual interactions may be much weaker. This concept has been tested by measuring the adhesion of platelets to collagen under conditions which could effect the formation of multiple interactions. Adhesion is markedly diminished at 4 °C but not at 23 or 37 oe. Metabolie inhibitors such as 2-deoxyglucose and Antimycin A do not inhibit adhesion although they virtually abolish subsequent aggregation. Brief formaldehyde fixation of platelets greatly reduces adhesion. These results are consistent with the concept that the formation of multiple linkcd interactions between the platelet surface and collagen are important in plateletcollagen adhesion and that mobility of platelet membrane components is required for the clustering of these interactions in focussed regions on the platelet surface. Key words: Adhesion, collagen, platelets
Introduction When the integrity of the vascular endothelium is disrupted, a platelet plug forms rapidly at the site of injury. Fibrillar collagen appears to be the most thrombogenic component of the vascular subendothelium (Baumgartner, 1977). Platelets first adhere to collagen. Adherent platelets then release their secretory granules, the contents of which result in the recruitment of other platelets to form the platelet plug. Several years aga it was demonstrated that before monomeric collagen couid initiate platelet aggregation, polymerization to collagen fibers was required (Muggli and Baumgartner, 1973; laffe and Deykin, 1974). Santoro and Cunning-
518
S. A. Santoro and L. W. Cunningham
harn (1977) suggested that this polymerization facilitated the development of multiple, simultaneous and linked interactions between the platelet surface and collagen which were required to initiate platelet secretion and subsequent aggregation. This concept has been supported by re cent demonstrations that ordered collagen polymers in which the arrangement of collagen monomers is different than in the native type fibril and polymerie collagen formed by randomly crosslinking monomeric collagen are all effective aggregating agents (Muggli, 1978; Santoro and Cunningham, 1980). The formation of multiple interactions between the platelet and collagen fiber could give rise to high affinity adhesion even if the individual interactions exhibit a much lower affinity (Santoro and Cunningham, 1977, 1980). In this communication we describe studies in which we have used a recently developed adhesion assay (Santoro and Cunningham, 1979) to examine the role of the mobility of platelet membrane components in the formation of multiple interactions between the platelet surface and collagen which result in high affinity adhesion. Materials and Methods
Collagen Human skin insoluble collagen was prepared as described by Cunningham and Ford (1968). It was suspended in 0.15 M NaCl, 0.05 M Tris-HCl, pH 7.4 at 2 mg/ ml and homogenized with a Potter Elvejhem tissue grinder.
Platelets Suspensions of washed, 51Cr-labeled platelets were prepared as previously described (Santoro and Cunningham, 1979). For use in the adhesion assay platelets were suspended in Dulbecco's phosphate-buffered saline supplemented with 5 mM glucose, 2 mM EDTA, and 0.3 % bovine serum albumin or they were resuspended in autologous plasma anticoagulated with 2 mM EDT A. Aggregation studies were performed with platelet rich plasma prepared as described earlier (Santoro and Cunningham, 1977) or with washed platelets prepared as above except that EDTA was omitted from the final resuspension media which was supplemented with 2mM CaCI 2 , 1 mM MgCl 2 and 0.05 % fibrinogen. Platelet concentrations were adjusted to 2.5 X lOB per ml.
Adhesion and Aggregation Assays Adhesion of platelets to insoluble collagen was determined in the same manner as adhesion to reconstituted collagen fibers (Santoro and Cunningham, 1979). Briefly, 0.4 ml of 51Cr-labeled platelets were incubated with 0.1 ml of the insoluble collagen homogenate containing the desired quantity of collagen in a reciprocal shaker operating at 100 cycles per min. Except for kinetic and temperature studies the incubation was for 20 min at 37 oe. The suspension was then applied to a 2.5 cm Unipore membrane (5 flm pore size) obtained from BioRad Laboratories mounted in a multiwell manifold which had been prerinsed with assay buffer. The membrane was then rinsed with two one ml portions of assay buffer. Radioactivity retained on the membrane was quantitated and the percentage of platelers in the incubation medium that adhered to collagen was determined.
Platelet-Collagen Adhesion
519
Platelet aggregation studies were performed in a Peyton Aggregometer as described earlier (Santoro and Cunningham, 1977).
Inhibition 01 Platelet Metabolism 33 mM 2-deoxyglucose (Sigma) was prepared in 0.15 M NaCI, 0.05 M TrisHCI, pH 7.4. 50 .ul of this solution was incubated with 0.4 ml of platelet suspension containing no added glucose for 30 min at room temperature. Antimycin A (Sigma) was dissolved in 95 Ofo ethanol at 10 mg/mI and then diluted to a concentration of 250 .ug/ml in 25 'll/o ethanol. 10 Jil of this solution was transferred to 0.4 ml platelet suspension and incubated for 30 min at room temperature. Exposure to this concentration of ethanol did not adversely affect platelet adhesion or aggregation.
Formaldehyde Fixation of Platelets Platelets were briefly fixed in the first was hing fluid after chromium labeling by incubation with 1/10 volume of 2 Ofo formaldehyde for 30 min at room temperature. The washing was then completed as for unfixed platelets.
Results
Adhesion Assay The use of the membrane-filtration assay to determine the adhesion of 51Cr_ labeled platelets to reconstituted native type collagen fibrils has been previously described (Santoro and Cunningham, 1979). This assay is shown here to be suitable for determining the adhesion of platelets to insoluble collagen isolated directly from skin. As shown in Figure 1 the extent of adhesion is nearly a linear function of the amount of collagen present in the assay. This finding is essentially identical to that of reconstituted collagen fibers (Santoro and Cunningham, 1979). However, higher concentrations of insoluble collagen are required to give levels of adhesion comparable to those obtained with reconstituted fibrils. This is likely to be because the finer suspension obtained upon homogenization of the reconstituted fibril gives a greater surface area on which adhesion may occur. The retention of platelets on the filter membrane is not due to trapping oi platelets within the collagen mat. In the absence of any collagen 0.6-2 percent of the platelets may be retained on the filter. If the platelets are incubated for 20 min in the presence of 200.ug of homogenized insoluble collagen and then transferred to the filter 30-40 percent of the platelets are retained. If, however, the collagen is first applied to the membrane and the platelets are added only 1-2 percent are retained. These results are also comparable to those obtained with reconstituted fibrils (Santoro and Cunningham, 1979).
Effect 01 Plasma To examine the possible role of plasma factors in the adhesion of platelets to collagen, washed, chromium-Iabeled platelets were resuspended in either autologous platelet-free plasma or a balanced, phosphate-buffered salt solution containing bovine serum albumin. The adhesion of platelets to insoluble collagen was
520
S. A. Santoro and L. W. Cunningham
40
c: 0
30
."
CI>
.r. '0
<{
c:
CI>
...
20
u
CI>
a..
10
50
100
150
200
Collagen, }Jg
Fig. 1. Dependence of adhesion assay on the concentration of insoluble collagen. Washed, labeled platelets (0.4 ml), 2.5 X 108 per ml were incubated for 20 min at 37°C with 0.1 ml of assay buffer containing the indicated amount of homogenized insoluble collagen. Adhesion is plotted as the percentage of total platelets in the incubation adhering to collagen. Error bars indicate the range of duplicate determinations. then determined. As shown in Figure 2, the adhesion in the presence of plasma was slightly less than but not significantly different from that occuring in the buffered salts-albumin solution. This finding does not eliminate the involvement of plasma factors in platelet-collagen adhesion, but does indicate that if such factors are involved they are bound sufficiently tightly to the platelet surface so that they are not removed by the washings employed in these studies. Kinetics and Temperature Effects The adhesion to collagen of platelets was measured as a function of time at 4, 23 and 37 oe. Platelets were preincubated for 5 minutes at the desired temperature before the addition of collagen to initiate the adhesion assay. The results of these studies are shown in Figure 3. At 37 oe the extent of adhesion increased rapidly with time until 5-10 minutes when near maximal levels were achieved. The results obtained at 23 oe were not markedly different. However at 4 oe the maximum adhesion was only about one third that obtained at 23 ° or 37 oe. Several different explanations are possible. These temperature effects may reflect thc thermodynamics of the interactions involved in platelet collagen adhesion, the need for active cellular metabolism to maintain the platelet surface in an adhesive state, or a requirementfor mobility ofmembrane components involved in adhesion, since membrane fluidity is greatly reduced at 4 oe.
Platelet-Collagen Adhesion
521
-r c
30 t-
r-- r--
-,-
0
-'--
'"
r--
Q)
-'=
r-
"0
«
....c
-
-'-
... Q)
0
Q)
0-
10 t-
p=-, a
b Plasma
a
~ b
PBS- albumin
Fig. 2. Effect of plasma on collagen-platelet adhesion. Washed, labeled platelets were resuspended in either autologous plasma or assay buffer at 2.5 X 108 platelets per ml. 0.4 ml of platelet suspension was incubated with 0.1 ml of 0.9 % NaCI or 0.9 % NaCI containing 300 flg of insoluble collagen for 20 min at 37 oe. The extent of adhesion was then determined. a. Platelets retained on the membrane after incubation with collagen. b. Platelets retained on the membrane after incubation in the absence of collagen. Error bars indicate the range of duplicate determinations.
Effect of Metabolie Inhibitors It has been demonstrated that agents such as 2-deoxyglucose or Antimycin A, an inhibitor of electron transport, which deplete the platelet metabolie pool, but not the secretable storage pool, of ATP are effective inhibitors of platelet aggregation induced by collagen and other agents (Murer et al., 1967; Murer, 1968; Kattlove and Gomez, 1975). These agents were employed to determine if the inhibition of adhesion which occurred at 4 oe might be due to the inhibition of platelet metabolism. Platelet suspensions were preincubated with 2-deoxyglucose or Antimycin A alone or in combination for 30 min at room temperature prior ro warming the suspension to 37 oe, and adding collagen to initiate the adhesion assay. As shown in Figure 4, 2-deoxyglucose (6 mM) or Antimycin A (5,ugiml) failed to produce significant inhibition of adhesion. The combination only reduced adhesion about 25 percent. In contrast aggregation initiated by 13 flg/ml collagen was inhibited 94 percent by 6 mM 2-deoxyglucose, 92 percent by 5 ,ug/ml Antimycin A and 99 per cent by the combination (data not shown). In contrast to the results obtained with metabolie inhibitors, markedly different results were obtained when platelets were briefly fixed with formaldehyde. As expected these platelets failed to aggregate in response to collagen. However, as
522
S. A. Santoro and L. W. Cunningham
o
c:
•
20
0
I/)
CI>
.c "0
c:
CI>
.... CI> a..
I.)
10 A
10
20
30
min. Fig. 3. Effect of temperature on collagen-platelet adhesion. 0.4 ml of platelets (2.5 X 108 per ml) was incubated with 0.1 ml of 0.9 Ofo NaCI containing 300 ,ug of insoluble collagen for the indicated times at 37 °C (0),23 °C (e), or 4 °C (A). Retention of platelets on the membrane in the absence of collagen at 4, 23 and 37 °C (6). Each point represents the mean of duplicate determinations.
30 c 0 ·in
.., CI)
.<:::
20
~
CI)
.,~ c..
10
abc d e Fig. 4. Effect of metabolic inhibitors on the adhesion of platelets to collagen. 0.4 ml of platelet suspension (2 X 108 per ml) containing the indicated concentration of 2-deoxyglucose or Antimycin A was incubated for 30 min at 23 oe. The suspension was brought to 37 °C and 0.1 ml of 0.9 Ofo NaCI or 0.9 Ofo NaCI containing 200,ug of insoluble collagen was added. The extent of adhesion was then determined in a 20 min assay at 37 degrees. a. Adhesion to 200 ,ug of collagen in the absence of metabolic inhibitors. b. Platelets retained on the membrane in the absence of collagen. c. Adhesion in the presence of 6 mM 2-deoxyglucose. d. Adhesion in the presence of 5 ,ug/ml Antimycin A. e. Adhesion in the presence of 6 mM 2-deoxyglucose and 5 ,ug/ml Antimycin A. Error bars represent the mean ± 1 SD of triplicate determinations.
Platelet-Collagen Adhesion
523
40
c
30
0
:ü;
'"
~
"0
C
20
'" u
~
a. '"
10
o
b
c
d
Fig. 5. Adhesion of formaldehyde-fixed platelets to insoluble collagen. Platelets were labeled and then incubated in the first washing solution or in washing solution containing 0.2 Ofo formaldehyde for 30 min. Platelets were again washed and resuspended in assay buffer. 004 ml of platelet suspension (2.5 X lOB per ml) was incubated with 0.1 ml of 0.9 Ofo NaCI or 0.9 Ofo NaCI containing 200 ftg of insoluble collagen. The extent of adhesion after 20 min was determined. a. Unfixed platelets plus 200 ftg collagen. b. Unfixed platelets in the absence of collagen. c. Formalin-fixed platelets in the presence of 200 ftg collagen. d. Formalin fixed platelets in the absence of collagen.
shown in Figure 5 the fixation also resulted in marked inhibition of adhesion. These platelets could be agglutinated by the antibiotic risticetin in the presence of plasma. Discussion Human platelets can be readily shown to adhere to insoluble collagen fibrils whether derived from skin or genera ted from purified soluble collagen. The requirement for interaction of platelets with insoluble multimers of collagen as a prelude to the release of granule contents and platelet aggregation is well established (Muggli and Baumgartner, 1973; Jaffe and Deykin, 1974; Brass and Bensllsan, 1974; Muggli, 1978; Santoro and Cunningham, 1980), but the initial development of such multiple linked interactions, adhesion, is not yet clearly understood. The nature of the platelet membrane component or components which are responsible for recognition and binding to collagen is not known. Several hypotheses have been put forward (Jamieson et al., 1971; Bensusan et al., 1978) but as yet none have been established (Santoro and Cunningham, 1977, 1979). The data presented here suggest that this platelet component is tightly membrane bound since binding ofwashed platelets to collagen is not dependentupon plasma. Such conclusions are, however, strongly dependent upon relatively ambiguom technical points such as platelet washing procedures and it cannot be ruled out 33
Collagen 1/6
524
S. A. Santoro and L. W. Cunningham
that tightly bound plasma components playa role in adhesion. For example, von Wille brand factor, which can be adsorbed by the genetically distinct collagens (Santoro, 1981), can be identified on the surface of platelets washed by the present procedure (Santoro, unpublished observation). Adhesion of a platelet to a collagen fibril is presumably initiated by a single interaction involving a site on the fibril and a "receptor" on'the platelet membrane. Functional adhesion leading to platelet aggregation requires the rapid subsequent development of multiple interactions of the same type or, conceivably, of more than this one type. It is of interest that inhibition of metabolism with 2-deoxvglucose and Antimycin A which deplete the platelet metabolie pool of A (Kattlove and Gomez, 1975) results in only a modest reduction in adhesion although platelet aggregation is markedly diminished. This result suggests that the development of multiple platelet-collagen interactions is not dependent on active metabolism. Additional insight is afforded by the measurements of platelet adhesion as a function of temperature. At 4 °C platelet adhesion is markedly diminished. This finding is in agreement with the re cent report of Mant (1980). In view of the effects noted with metabolie inhibitors, it seems unlikely that this decrease is the result of diminished metabolism. Another possible explanation may lie in the restricted mobility of membrane components at this temperature because of increased viscosity or, possibly, a lipid phase transition (Melchior and Steim, 1976). This restricted mobility of membrane components, including collagen receptors, could delay or prevent the formation of the multiple collagen-platelet interactions required for high affinity, aggregation-linked adhesion. The effects no ted with formaldehyde-fixed platelets are consistent with this interpretation. Exposure to formaldehyde cross-links membrane proteins and rest riets their mobility in the membrane. Formaldehyde-treated platelets show greatly reduced ability to adhere to collagen fibrils, again supporting an inability to form multiple sites of adhesion. A common feature of these inhibitory processes lies in their restrietion of mobility of receptors within the platelet membrane. Formation of multiple interactions between the platelet and collagen fibril requires the movement of receptors in the membrane to positions which permit interactions :1t intervals along the collagen fibril. The fibril sites are rigidly fixed, and the fiber may be considered thus to be inducing organization, "clustering", of receptors on the platelet surface. This model is illustrated in schematic form in Figure 6. The initial complex formed between a single platelet surface receptor and collagen is of relatively low affinity and is readily disrupted by was hing and similar shear forces genera ted in the assay. Diffusion of additional receptors within the fluid platelet membrane permits the rapid development of a cluster of additional interactions, which result, overall, in high affinity adhesion (Santoro and Cunningham, 1977, 1980). Any treatment which reduces membrane fluidity limits the platelet collagen interaction to only single interaction, low affinity adhesion which is not detected by this assay and is non-productive in terms of platelet aggregation. Clustering of receptors following interaction with hormones, lipoproteins, lectins, a2 macroglobulin, antibodies and other agents is a weIl known prelude to formation of coated pits and internalization by the process of adsorptive pinocytosis (Goldstein et al., 1979). This process appears to involve coordinated movement of intracellular elements. It is possible that clustering of receptors induced by a geometrically constrained, multivalent extern al agent such as a collagen fibril may activate a related process which results in active secretion
TP
Platelet-Collagen Adhesion
B.
A.
525
c.
Fig. 6. Schematic diagram indicating the possible role of and adhesion site clustering in the development of high affinity adhesion. Prior to clustering only low affinity adhesion exists (A, B). Clustering facilitates the formation of multiple, simultaneous interactions (C) giving rise to high affinity adhesion. This latter event is inhibited by formaldehyde fixation but not by metabolie inhibitors.
of intracellular components from platelet granules by fusion with the plasma membrane. Platelet release and aggregation stimulated by antigen-antibody complexes appear to result from a similar process (Henson and Spiegelberg, 1973).
Acknowledgments We acknowledge the able technical asistance of Mr. Joseph F. Cowan and the support of a USPHS Biomedical Research Support Grant to Vanderbilt University School of Medicine.
References Baumgartner, H. R.: Platelet interaction with collagen fibrils in flowing blood. 1. Reaction of human platelets with a-chymotrypsin-digested subendothelium. Thromb. Haemostas. 37: 1-16,1977. Bensusan, H. B., Koh, T. L., Henry, K. G., Murray, B. A. and Culp, L. A.: Evidence that fibronectin is the collagen receptor on platelet membranes. Proc. Natl. Acad. Sei. U.S.A. 75: 5864-5868, 1978. Brass, T. F. and Bensusan, H. B.: The role of collagen quaternary structure in the plateletcollagen interaction. J. Clin. Invest. 54: 1480-1487, 1974. Cunningham, L. W. and f'ord, J. D.: A comparison of the glycopeptides derived from soluble and insoluble collagens. J. Biol. Chern. 253: 2390-2398, 1968. Goldstein, J. L., Anderson, R. G. and Brown, M. S.: Coated pits, coated vesicles and receptor-mediated endocytosis. Nature 279: 679-685, 1979. Henson, P. M. and Spiegelberg, H. L.: The release of serotonin from human platelets induced by aggregated immunoglobulins of different classes and subclasses. J. Clin. Invest. 52: 1282-1288, 1973. Jaffe, R. and Deykin, D.: Evidence for a structural requirement for the aggregation of platelets by collagen. J. Clin. Invest. 53: 875-883, 1974. Jamieson, G. A., Urban, C. L. and Barber, A. J.: Enzymatic basis for platelet-collagen adhesion as the primary step in haemostasis. Nat. New Biol. 234: 5-7, 1971. Kattlove, H. and Gomez, M. H.: Collagen-induced platelet aggregation: The role of
526
S. A. Santoro and L. W. Cunningham
adenine nucleotides and the release reaction. Thromb. Diath. Haemorrh. 34: 795-805, 1975. Mant, M. J.: Platelet adherence to collagen: Metabolie energy requirements. Thromb. Res. 17: 729-736, 1980. Melchior, D. L. and Steim, J. M.: Thermotropic transitions in biomembranes. Ann. Rev. Biophys. Bioeng. 5: 205-238, 1976. Muggli, R.: Collagen induced platelet aggregation: Native collagen quaternary structure is not an essential structural requirement. Thromb. Res. 13: 829-843, 1978. Muggli, R. and Baumgartner, H. R.: Collagen induced platelet aggregation: Requirement for tropocollagen multimers. Thromb. Res. 3: 715-728, 1973. Murer, E. H.: Release reaction and energy metabolism in blood platelets with special reference to the burst in oxygen uptake. Bioehim. Biophys. Acta 162: 320-326, 1968. Murer, E. H., Hellem, A. J. and Rozenberg, M. c.: Energy metabolism and platelet function. Seand. J. Clin. Lab. Invest. 19: 280-282, 1967. Santoro, S. A.: Adsorption of von Willebrand factor/factor VIII by the genetically distinct interstitial collagens. Thromb. Res. 21: 689-693, 1981. Santoro, S. A. and Cunningham, L. W.: Collagen mediated platelet aggregation: Evidence for multivalent interactions of intermediate specificity between collagen and platelets. J. Clin. Invest. 60: 1054-1060, 1977. Santoro, S. A. and Cunningham, L. W.: Fibronectin and the multible inter action model for platelet-collagen adhesion. Proe. Natl. Aead. Sei. U. S. A. 76: 2644-2648, 1979. Santoro, S. A. and Cunningham, L. W.: Collagen mediated platelet aggregation: The role of multiple interactions between the platelet surface and collagen. Thromb. Haemostos. 43: 158-162,1980. Dr. Leon W. Cunningham, Department of Biochemistry, Vanderbilt University, School of Medicine, Nashville, TN 37232, USA.
Platelet-Collagen Adhesion - Membrane Fluidity and the Development of High Affinity Adhesion through Multiple Interacting Sites SAMUEL A. SANTOROl and LEON
w.
CUNNINGHAM2
Division of Laboratory Medicine, Washington University Schoo! of Medicine, St. Louis, MO 63110, USA 2 Department of Biochemistry, Vanderbilt University Schoo! of Medicine, Nashville, TN 37232, USA 1
Abstract Multiple, linked interactions between the platelet surface and collagen fibers have been implicated in the initiation of platelet secretion and subsequent aggregation. The formation of such multiple simultaneous interactions could give rise to high affinity adhesion of platelets to collagen even though the affinity of thc individual interactions may be much weaker. This concept has been tested by measuring the adhesion of platelets to collagen under conditions which could effect the formation of multiple interactions. Adhesion is markedly diminished at 4 °C but not at 23 or 37 oe. Metabolie inhibitors such as 2-deoxyglucose and Antimycin A do not inhibit adhesion although they virtually abolish subsequent aggregation. Brief formaldehyde fixation of platelets greatly reduces adhesion. These results are consistent with the concept that the formation of multiple linkcd interactions between the platelet surface and collagen are important in plateletcollagen adhesion and that mobility of platelet membrane components is required for the clustering of these interactions in focussed regions on the platelet surface. Key words: Adhesion, collagen, platelets
Introduction When the integrity of the vascular endothelium is disrupted, a platelet plug forms rapidly at the site of injury. Fibrillar collagen appears to be the most thrombogenic component of the vascular subendothelium (Baumgartner, 1977). Platelets first adhere to collagen. Adherent platelets then release their secretory granules, the contents of which result in the recruitment of other platelets to form the platelet plug. Several years aga it was demonstrated that before monomeric collagen couid initiate platelet aggregation, polymerization to collagen fibers was required (Muggli and Baumgartner, 1973; laffe and Deykin, 1974). Santoro and Cunning-
518
S. A. Santoro and L. W. Cunningham
harn (1977) suggested that this polymerization facilitated the development of multiple, simultaneous and linked interactions between the platelet surface and collagen which were required to initiate platelet secretion and subsequent aggregation. This concept has been supported by re cent demonstrations that ordered collagen polymers in which the arrangement of collagen monomers is different than in the native type fibril and polymerie collagen formed by randomly crosslinking monomeric collagen are all effective aggregating agents (Muggli, 1978; Santoro and Cunningham, 1980). The formation of multiple interactions between the platelet and collagen fiber could give rise to high affinity adhesion even if the individual interactions exhibit a much lower affinity (Santoro and Cunningham, 1977, 1980). In this communication we describe studies in which we have used a recently developed adhesion assay (Santoro and Cunningham, 1979) to examine the role of the mobility of platelet membrane components in the formation of multiple interactions between the platelet surface and collagen which result in high affinity adhesion. Materials and Methods
Collagen Human skin insoluble collagen was prepared as described by Cunningham and Ford (1968). It was suspended in 0.15 M NaCl, 0.05 M Tris-HCl, pH 7.4 at 2 mg/ ml and homogenized with a Potter Elvejhem tissue grinder.
Platelets Suspensions of washed, 51Cr-labeled platelets were prepared as previously described (Santoro and Cunningham, 1979). For use in the adhesion assay platelets were suspended in Dulbecco's phosphate-buffered saline supplemented with 5 mM glucose, 2 mM EDTA, and 0.3 % bovine serum albumin or they were resuspended in autologous plasma anticoagulated with 2 mM EDT A. Aggregation studies were performed with platelet rich plasma prepared as described earlier (Santoro and Cunningham, 1977) or with washed platelets prepared as above except that EDTA was omitted from the final resuspension media which was supplemented with 2mM CaCI 2 , 1 mM MgCl 2 and 0.05 % fibrinogen. Platelet concentrations were adjusted to 2.5 X lOB per ml.
Adhesion and Aggregation Assays Adhesion of platelets to insoluble collagen was determined in the same manner as adhesion to reconstituted collagen fibers (Santoro and Cunningham, 1979). Briefly, 0.4 ml of 51Cr-labeled platelets were incubated with 0.1 ml of the insoluble collagen homogenate containing the desired quantity of collagen in a reciprocal shaker operating at 100 cycles per min. Except for kinetic and temperature studies the incubation was for 20 min at 37 oe. The suspension was then applied to a 2.5 cm Unipore membrane (5 flm pore size) obtained from BioRad Laboratories mounted in a multiwell manifold which had been prerinsed with assay buffer. The membrane was then rinsed with two one ml portions of assay buffer. Radioactivity retained on the membrane was quantitated and the percentage of platelers in the incubation medium that adhered to collagen was determined.
Platelet-Collagen Adhesion
519
Platelet aggregation studies were performed in a Peyton Aggregometer as described earlier (Santoro and Cunningham, 1977).
Inhibition 01 Platelet Metabolism 33 mM 2-deoxyglucose (Sigma) was prepared in 0.15 M NaCI, 0.05 M TrisHCI, pH 7.4. 50 .ul of this solution was incubated with 0.4 ml of platelet suspension containing no added glucose for 30 min at room temperature. Antimycin A (Sigma) was dissolved in 95 Ofo ethanol at 10 mg/mI and then diluted to a concentration of 250 .ug/ml in 25 'll/o ethanol. 10 Jil of this solution was transferred to 0.4 ml platelet suspension and incubated for 30 min at room temperature. Exposure to this concentration of ethanol did not adversely affect platelet adhesion or aggregation.
Formaldehyde Fixation of Platelets Platelets were briefly fixed in the first was hing fluid after chromium labeling by incubation with 1/10 volume of 2 Ofo formaldehyde for 30 min at room temperature. The washing was then completed as for unfixed platelets.
Results
Adhesion Assay The use of the membrane-filtration assay to determine the adhesion of 51Cr_ labeled platelets to reconstituted native type collagen fibrils has been previously described (Santoro and Cunningham, 1979). This assay is shown here to be suitable for determining the adhesion of platelets to insoluble collagen isolated directly from skin. As shown in Figure 1 the extent of adhesion is nearly a linear function of the amount of collagen present in the assay. This finding is essentially identical to that of reconstituted collagen fibers (Santoro and Cunningham, 1979). However, higher concentrations of insoluble collagen are required to give levels of adhesion comparable to those obtained with reconstituted fibrils. This is likely to be because the finer suspension obtained upon homogenization of the reconstituted fibril gives a greater surface area on which adhesion may occur. The retention of platelets on the filter membrane is not due to trapping oi platelets within the collagen mat. In the absence of any collagen 0.6-2 percent of the platelets may be retained on the filter. If the platelets are incubated for 20 min in the presence of 200.ug of homogenized insoluble collagen and then transferred to the filter 30-40 percent of the platelets are retained. If, however, the collagen is first applied to the membrane and the platelets are added only 1-2 percent are retained. These results are also comparable to those obtained with reconstituted fibrils (Santoro and Cunningham, 1979).
Effect 01 Plasma To examine the possible role of plasma factors in the adhesion of platelets to collagen, washed, chromium-Iabeled platelets were resuspended in either autologous platelet-free plasma or a balanced, phosphate-buffered salt solution containing bovine serum albumin. The adhesion of platelets to insoluble collagen was
520
S. A. Santoro and L. W. Cunningham
40
c: 0
30
."
CI>
.r. '0
<{
c:
CI>
...
20
u
CI>
a..
10
50
100
150
200
Collagen, }Jg
Fig. 1. Dependence of adhesion assay on the concentration of insoluble collagen. Washed, labeled platelets (0.4 ml), 2.5 X 108 per ml were incubated for 20 min at 37°C with 0.1 ml of assay buffer containing the indicated amount of homogenized insoluble collagen. Adhesion is plotted as the percentage of total platelets in the incubation adhering to collagen. Error bars indicate the range of duplicate determinations. then determined. As shown in Figure 2, the adhesion in the presence of plasma was slightly less than but not significantly different from that occuring in the buffered salts-albumin solution. This finding does not eliminate the involvement of plasma factors in platelet-collagen adhesion, but does indicate that if such factors are involved they are bound sufficiently tightly to the platelet surface so that they are not removed by the washings employed in these studies. Kinetics and Temperature Effects The adhesion to collagen of platelets was measured as a function of time at 4, 23 and 37 oe. Platelets were preincubated for 5 minutes at the desired temperature before the addition of collagen to initiate the adhesion assay. The results of these studies are shown in Figure 3. At 37 oe the extent of adhesion increased rapidly with time until 5-10 minutes when near maximal levels were achieved. The results obtained at 23 oe were not markedly different. However at 4 oe the maximum adhesion was only about one third that obtained at 23 ° or 37 oe. Several different explanations are possible. These temperature effects may reflect thc thermodynamics of the interactions involved in platelet collagen adhesion, the need for active cellular metabolism to maintain the platelet surface in an adhesive state, or a requirementfor mobility ofmembrane components involved in adhesion, since membrane fluidity is greatly reduced at 4 oe.
Platelet-Collagen Adhesion
521
-r c
30 t-
r-- r--
-,-
0
-'--
'"
r--
Q)
-'=
r-
"0
«
....c
-
-'-
... Q)
0
Q)
0-
10 t-
p=-, a
b Plasma
a
~ b
PBS- albumin
Fig. 2. Effect of plasma on collagen-platelet adhesion. Washed, labeled platelets were resuspended in either autologous plasma or assay buffer at 2.5 X 108 platelets per ml. 0.4 ml of platelet suspension was incubated with 0.1 ml of 0.9 % NaCI or 0.9 % NaCI containing 300 flg of insoluble collagen for 20 min at 37 oe. The extent of adhesion was then determined. a. Platelets retained on the membrane after incubation with collagen. b. Platelets retained on the membrane after incubation in the absence of collagen. Error bars indicate the range of duplicate determinations.
Effect of Metabolie Inhibitors It has been demonstrated that agents such as 2-deoxyglucose or Antimycin A, an inhibitor of electron transport, which deplete the platelet metabolie pool, but not the secretable storage pool, of ATP are effective inhibitors of platelet aggregation induced by collagen and other agents (Murer et al., 1967; Murer, 1968; Kattlove and Gomez, 1975). These agents were employed to determine if the inhibition of adhesion which occurred at 4 oe might be due to the inhibition of platelet metabolism. Platelet suspensions were preincubated with 2-deoxyglucose or Antimycin A alone or in combination for 30 min at room temperature prior ro warming the suspension to 37 oe, and adding collagen to initiate the adhesion assay. As shown in Figure 4, 2-deoxyglucose (6 mM) or Antimycin A (5,ugiml) failed to produce significant inhibition of adhesion. The combination only reduced adhesion about 25 percent. In contrast aggregation initiated by 13 flg/ml collagen was inhibited 94 percent by 6 mM 2-deoxyglucose, 92 percent by 5 ,ug/ml Antimycin A and 99 per cent by the combination (data not shown). In contrast to the results obtained with metabolie inhibitors, markedly different results were obtained when platelets were briefly fixed with formaldehyde. As expected these platelets failed to aggregate in response to collagen. However, as
522
S. A. Santoro and L. W. Cunningham
o
c:
•
20
0
I/)
CI>
.c "0
c:
CI>
.... CI> a..
I.)
10 A
10
20
30
min. Fig. 3. Effect of temperature on collagen-platelet adhesion. 0.4 ml of platelets (2.5 X 108 per ml) was incubated with 0.1 ml of 0.9 Ofo NaCI containing 300 ,ug of insoluble collagen for the indicated times at 37 °C (0),23 °C (e), or 4 °C (A). Retention of platelets on the membrane in the absence of collagen at 4, 23 and 37 °C (6). Each point represents the mean of duplicate determinations.
30 c 0 ·in
.., CI)
.<:::
20
~
CI)
.,~ c..
10
abc d e Fig. 4. Effect of metabolic inhibitors on the adhesion of platelets to collagen. 0.4 ml of platelet suspension (2 X 108 per ml) containing the indicated concentration of 2-deoxyglucose or Antimycin A was incubated for 30 min at 23 oe. The suspension was brought to 37 °C and 0.1 ml of 0.9 Ofo NaCI or 0.9 Ofo NaCI containing 200,ug of insoluble collagen was added. The extent of adhesion was then determined in a 20 min assay at 37 degrees. a. Adhesion to 200 ,ug of collagen in the absence of metabolic inhibitors. b. Platelets retained on the membrane in the absence of collagen. c. Adhesion in the presence of 6 mM 2-deoxyglucose. d. Adhesion in the presence of 5 ,ug/ml Antimycin A. e. Adhesion in the presence of 6 mM 2-deoxyglucose and 5 ,ug/ml Antimycin A. Error bars represent the mean ± 1 SD of triplicate determinations.
Platelet-Collagen Adhesion
523
40
c
30
0
:ü;
'"
~
"0
C
20
'" u
~
a. '"
10
o
b
c
d
Fig. 5. Adhesion of formaldehyde-fixed platelets to insoluble collagen. Platelets were labeled and then incubated in the first washing solution or in washing solution containing 0.2 Ofo formaldehyde for 30 min. Platelets were again washed and resuspended in assay buffer. 004 ml of platelet suspension (2.5 X lOB per ml) was incubated with 0.1 ml of 0.9 Ofo NaCI or 0.9 Ofo NaCI containing 200 ftg of insoluble collagen. The extent of adhesion after 20 min was determined. a. Unfixed platelets plus 200 ftg collagen. b. Unfixed platelets in the absence of collagen. c. Formalin-fixed platelets in the presence of 200 ftg collagen. d. Formalin fixed platelets in the absence of collagen.
shown in Figure 5 the fixation also resulted in marked inhibition of adhesion. These platelets could be agglutinated by the antibiotic risticetin in the presence of plasma. Discussion Human platelets can be readily shown to adhere to insoluble collagen fibrils whether derived from skin or genera ted from purified soluble collagen. The requirement for interaction of platelets with insoluble multimers of collagen as a prelude to the release of granule contents and platelet aggregation is well established (Muggli and Baumgartner, 1973; Jaffe and Deykin, 1974; Brass and Bensllsan, 1974; Muggli, 1978; Santoro and Cunningham, 1980), but the initial development of such multiple linked interactions, adhesion, is not yet clearly understood. The nature of the platelet membrane component or components which are responsible for recognition and binding to collagen is not known. Several hypotheses have been put forward (Jamieson et al., 1971; Bensusan et al., 1978) but as yet none have been established (Santoro and Cunningham, 1977, 1979). The data presented here suggest that this platelet component is tightly membrane bound since binding ofwashed platelets to collagen is not dependentupon plasma. Such conclusions are, however, strongly dependent upon relatively ambiguom technical points such as platelet washing procedures and it cannot be ruled out 33
Collagen 1/6
524
S. A. Santoro and L. W. Cunningham
that tightly bound plasma components playa role in adhesion. For example, von Wille brand factor, which can be adsorbed by the genetically distinct collagens (Santoro, 1981), can be identified on the surface of platelets washed by the present procedure (Santoro, unpublished observation). Adhesion of a platelet to a collagen fibril is presumably initiated by a single interaction involving a site on the fibril and a "receptor" on'the platelet membrane. Functional adhesion leading to platelet aggregation requires the rapid subsequent development of multiple interactions of the same type or, conceivably, of more than this one type. It is of interest that inhibition of metabolism with 2-deoxvglucose and Antimycin A which deplete the platelet metabolie pool of A (Kattlove and Gomez, 1975) results in only a modest reduction in adhesion although platelet aggregation is markedly diminished. This result suggests that the development of multiple platelet-collagen interactions is not dependent on active metabolism. Additional insight is afforded by the measurements of platelet adhesion as a function of temperature. At 4 °C platelet adhesion is markedly diminished. This finding is in agreement with the re cent report of Mant (1980). In view of the effects noted with metabolie inhibitors, it seems unlikely that this decrease is the result of diminished metabolism. Another possible explanation may lie in the restricted mobility of membrane components at this temperature because of increased viscosity or, possibly, a lipid phase transition (Melchior and Steim, 1976). This restricted mobility of membrane components, including collagen receptors, could delay or prevent the formation of the multiple collagen-platelet interactions required for high affinity, aggregation-linked adhesion. The effects no ted with formaldehyde-fixed platelets are consistent with this interpretation. Exposure to formaldehyde cross-links membrane proteins and rest riets their mobility in the membrane. Formaldehyde-treated platelets show greatly reduced ability to adhere to collagen fibrils, again supporting an inability to form multiple sites of adhesion. A common feature of these inhibitory processes lies in their restrietion of mobility of receptors within the platelet membrane. Formation of multiple interactions between the platelet and collagen fibril requires the movement of receptors in the membrane to positions which permit interactions :1t intervals along the collagen fibril. The fibril sites are rigidly fixed, and the fiber may be considered thus to be inducing organization, "clustering", of receptors on the platelet surface. This model is illustrated in schematic form in Figure 6. The initial complex formed between a single platelet surface receptor and collagen is of relatively low affinity and is readily disrupted by was hing and similar shear forces genera ted in the assay. Diffusion of additional receptors within the fluid platelet membrane permits the rapid development of a cluster of additional interactions, which result, overall, in high affinity adhesion (Santoro and Cunningham, 1977, 1980). Any treatment which reduces membrane fluidity limits the platelet collagen interaction to only single interaction, low affinity adhesion which is not detected by this assay and is non-productive in terms of platelet aggregation. Clustering of receptors following interaction with hormones, lipoproteins, lectins, a2 macroglobulin, antibodies and other agents is a weIl known prelude to formation of coated pits and internalization by the process of adsorptive pinocytosis (Goldstein et al., 1979). This process appears to involve coordinated movement of intracellular elements. It is possible that clustering of receptors induced by a geometrically constrained, multivalent extern al agent such as a collagen fibril may activate a related process which results in active secretion
TP
Platelet-Collagen Adhesion
B.
A.
525
c.
Fig. 6. Schematic diagram indicating the possible role of and adhesion site clustering in the development of high affinity adhesion. Prior to clustering only low affinity adhesion exists (A, B). Clustering facilitates the formation of multiple, simultaneous interactions (C) giving rise to high affinity adhesion. This latter event is inhibited by formaldehyde fixation but not by metabolie inhibitors.
of intracellular components from platelet granules by fusion with the plasma membrane. Platelet release and aggregation stimulated by antigen-antibody complexes appear to result from a similar process (Henson and Spiegelberg, 1973).
Acknowledgments We acknowledge the able technical asistance of Mr. Joseph F. Cowan and the support of a USPHS Biomedical Research Support Grant to Vanderbilt University School of Medicine.
References Baumgartner, H. R.: Platelet interaction with collagen fibrils in flowing blood. 1. Reaction of human platelets with a-chymotrypsin-digested subendothelium. Thromb. Haemostas. 37: 1-16,1977. Bensusan, H. B., Koh, T. L., Henry, K. G., Murray, B. A. and Culp, L. A.: Evidence that fibronectin is the collagen receptor on platelet membranes. Proc. Natl. Acad. Sei. U.S.A. 75: 5864-5868, 1978. Brass, T. F. and Bensusan, H. B.: The role of collagen quaternary structure in the plateletcollagen interaction. J. Clin. Invest. 54: 1480-1487, 1974. Cunningham, L. W. and f'ord, J. D.: A comparison of the glycopeptides derived from soluble and insoluble collagens. J. Biol. Chern. 253: 2390-2398, 1968. Goldstein, J. L., Anderson, R. G. and Brown, M. S.: Coated pits, coated vesicles and receptor-mediated endocytosis. Nature 279: 679-685, 1979. Henson, P. M. and Spiegelberg, H. L.: The release of serotonin from human platelets induced by aggregated immunoglobulins of different classes and subclasses. J. Clin. Invest. 52: 1282-1288, 1973. Jaffe, R. and Deykin, D.: Evidence for a structural requirement for the aggregation of platelets by collagen. J. Clin. Invest. 53: 875-883, 1974. Jamieson, G. A., Urban, C. L. and Barber, A. J.: Enzymatic basis for platelet-collagen adhesion as the primary step in haemostasis. Nat. New Biol. 234: 5-7, 1971. Kattlove, H. and Gomez, M. H.: Collagen-induced platelet aggregation: The role of
526
S. A. Santoro and L. W. Cunningham
adenine nucleotides and the release reaction. Thromb. Diath. Haemorrh. 34: 795-805, 1975. Mant, M. J.: Platelet adherence to collagen: Metabolie energy requirements. Thromb. Res. 17: 729-736, 1980. Melchior, D. L. and Steim, J. M.: Thermotropic transitions in biomembranes. Ann. Rev. Biophys. Bioeng. 5: 205-238, 1976. Muggli, R.: Collagen induced platelet aggregation: Native collagen quaternary structure is not an essential structural requirement. Thromb. Res. 13: 829-843, 1978. Muggli, R. and Baumgartner, H. R.: Collagen induced platelet aggregation: Requirement for tropocollagen multimers. Thromb. Res. 3: 715-728, 1973. Murer, E. H.: Release reaction and energy metabolism in blood platelets with special reference to the burst in oxygen uptake. Bioehim. Biophys. Acta 162: 320-326, 1968. Murer, E. H., Hellem, A. J. and Rozenberg, M. c.: Energy metabolism and platelet function. Seand. J. Clin. Lab. Invest. 19: 280-282, 1967. Santoro, S. A.: Adsorption of von Willebrand factor/factor VIII by the genetically distinct interstitial collagens. Thromb. Res. 21: 689-693, 1981. Santoro, S. A. and Cunningham, L. W.: Collagen mediated platelet aggregation: Evidence for multivalent interactions of intermediate specificity between collagen and platelets. J. Clin. Invest. 60: 1054-1060, 1977. Santoro, S. A. and Cunningham, L. W.: Fibronectin and the multible inter action model for platelet-collagen adhesion. Proe. Natl. Aead. Sei. U. S. A. 76: 2644-2648, 1979. Santoro, S. A. and Cunningham, L. W.: Collagen mediated platelet aggregation: The role of multiple interactions between the platelet surface and collagen. Thromb. Haemostos. 43: 158-162,1980. Dr. Leon W. Cunningham, Department of Biochemistry, Vanderbilt University, School of Medicine, Nashville, TN 37232, USA.