J. BIOMED. MATER. RES.
VOL. 11, PP. 365-372 (1977)
Inhibition and Enhancement of Platelet Aggregation by Collagen Derivatives W. W. KAY,* E. KURYLO, G. CHONG, and B. BHARADWAJ, Department of Surgery, University of Saskatchewan, Saskatoon, Saskatchewan, Canada, S7N OW0
Summary Rat tail tendon tropocollagen rapidly caused the aggregation of blood platelets coincident with collagen multimerization. Several long-chain alkyl derivatives of collagen] however, were completely unable to induce platelet aggregation even at unusually high concentrations. These collagen derivatives varied in their polymerization properties, and the polymerization of tropocollagen derivatives was shown not necessarily to be a prerequisite for platelet aggregation. These collagen derivatives may enhance the antithrombogenic properties of collagenbaaed prostheses. A short-chain derivative, collagen-glycine ethyl ester, profoundly enhanced platelet aggregation in the absence of any measurable polymerization. This derivative may be useful as a growth-promoting burn dressing.
INTRODUCTION Collagen is the most abundant and ubiquitous animal protein, largely functioning as supporting structures for various tissues. This unique protein is generally present in four genetically distinct types coded for by a t least five structural genes.' The evolution of such specialized proteins provides a strong, malleable, and chemically versatile molecule in abundance which is amendable to chemical modification and, as such, is a valuable material for the construction of biomedical prostheses. Indeed, various collagen polymers and copolymers are proving t o be highly advantageous alternatives t o more conventional medical pros these^,*-^ wound dressing^,^-^ and dialysis membranes. s-lo * Present address : Department of Bacteriology and Biochemistry] University of Victoria, Victoria, B.C. Canada VSW 2Y2. 365 @ 1977 by John Wiley & Sons, Inc.
366
KAY ET AL.
One of the major drawbacks to the use of various collagen prosthetic devises is the biological incompatability with respect to blood platelet adhesion. 11-15 Since we have recently reported that modified facia lata tricuspid valves of free-floating atrial grafts sustained unusually high longevity in dogs, we sought to assess the degree of antithrombogenicity of our grafts using the platelet-collagen aggregation response as an indicator of thrombogenic activity. To this end we described herein the preparation and properties of, as well as platelet responses to, various chemically modified tropocollagens which are either highly thrombogenic or antithrombogenic.
MATERIALS AND METHODS Rat tail tendon collagen was prepared by acetic acid extraction,15 exhaustive dialysis, and lyophilization. Glycine ethyl ester was coupled t o collagen carboxyl groups with a water-soluble carbodiimide [l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide]16 and the degree of modification estimated with 14C-glycineethyl ester (1.62 mmol/g collagen). Either oleylamine or cetylamine (66.5 mM) was crosslinked to collagen in 95y0 ethyl alcohol with dicyclohexylcarbodiimide (88.7 mM), and the degree of crosslinking estimated with 3H-octadecylamine (New England Nuclear, custom tritiation). Oleylamine-collagen-glycine ethyl ester was prepared after reaction with oleylamine by similar methods. Succinylated collagen was prepared by reacting collagen with 1 M succinic anhydride in 0.1 M NaHC03. Succinic anhydride was added stepwise a t 0°C over a period of 4 hr. After 12 hr, succinic anhydride was again added (0.5 M ) and was reacted for an additional 24 hr a t room temperature, dialyzcd, Iyophilized, and redisolved a t pH 8.0. Acetylated collagen was prepared by reacting collagen in saturated NaHCO, with 0.5 M acetic anhydride for 19 hr at 0°C. Trinotrobenzene collagen was prepared by reaction of rat tail collagen with trinotrobenzenesulfonic acid (1.66 mM) in 20/, acetic acid for 24 hr a t room temperature. All derivatives werp exhaustively dialyzed and Iyophilized prior to use. Blood was collected from the jugular vcin of a dog and mixed immediately with ACD (1:6 v/v). Platelet-rich plasma for aggregation studies was obtaincd after centrifugation a t 120 g’s for 20 min. Aggrpgation was measured on a T’eyton aggr:grcyyncltcressentially by the method of Born.16
COLLAGEN DERIVATIVES AND PLATELET AGGREGATION 367
Collagen polymerization was measured by reflected 420 nm light on a n Aminco microspectrophotometer operated as a nephelometer. Polymerization was followed at 37°C in 0.18 M Tris HCl (pH 7.3). Platelet-collagen interactions were followed directly by scanning electron microscopy on a Cambridge stereoscanning electron microscope. Platelets in plasma were fixed with 1.7% glutaraldehyde, caught on Nucleopore membrane filters, dehydrated in isoamyl alcohol, and critical point-dried. The samples were then plated with gold.
RESULTS AND DISCUSSION All collagen derivatives used in this study were primarily blocked a t free amino or carboxyl groups and, as predicted,” this should interfere with polymerization by disrupting charge group interactions. Clearly, these procedures largely inhibited collagen polymerization (Table I), with the exception of oleyl collagen which could still polymerize, albeit incompletely. Commensurate with the inhibition of polymerization, five of the seven collagen derivatives were subsequently unable to initiate platelet aggregation; however, both glycine ethyl ester and the oleyl-glycine ethyl ester derivatives were TABLE I Platelet Aggregation by Rat Tail Tropocollagen Derivatives
Collagen Derivative Native collagen Glyeine ethyl ester-collagen
Oleyl collagen Oleyl-glycine ethyl ester Cetyl/collagen Succinyl/collagen Acetyl/collagen Trinitrobenzenesulfonyl collagen
Concentration (rg/ml) 100 100 5.0 0.5 500 220 140 300 200 230
Aggregation Time. (set) 120 10 10
28 >20 hr 60 >20 hr >20 hr >20 hr >20 hr
Polymerization (relative intensity)b 100 0 0 0 48 0 15 12 0 12
* Defined as the time elapsed between the addition of collagen and the increase in light transmittance at the beginning of platelet aggregation. b Polymerization was estimated in relative intensity units with rat tail collagen at 100 pg/ml at 10 units. Maximum polymerization in 2 hr was recorded.
KAY ET AL.
368
still potent platelet aggregating agents, particularly glycine ethyl ester-collagen which exhibited a 200-fold increase in aggregating potency relative to native rat tail collagen. Collagen polymerization was found to occur in plasma as well under these conditions. Scanning electron microscopy readily showed extensive fiber formation. I n addition, 1251-collagenscan be readily sedimented by ultracentrifugation (unpublished results). Typical aggregation curves for some of these derivatives are shown in Figure 1. The immediate and complete aggregation by glycine ethyl ester-collagen is compared to the time-dependent aggregation of rat, tail tropocollagen. The
W
0
z
4 + !z
5
tn z 4
K
c W
> F
5W K
I-
1 min
Fig. 1. Aggregation of canine platelets by native collagen and modified collagens: (A) collagen-glycine ethyl ester (5 fig/ml) ; (B) oleyl-collagen-glycine ethyl ester (110 pg/ml); (C) native acid soluble rat tail tropocollagen (212 pg/ml); (D) Acyl-collagen (200 pg/ml). No changes were found with oleyl- or cetylcollagens.
COLLAGEN DERIVATIVES A N D PLATELET AGGREGATION 369
time dependence is thought by some investigators to represent p o l y m e r i ~ a t i o n , 'although ~ ~ ~ ~ from the results of Table I, the nonpolymerizing derivative collagen-glycine ethyl ester is an unusually potent aggregating agent by a t least an order of magnitude greater than any collagen that we are aware of. This may possibly be due to the creation of a basic collagen since the aid groups have been blocked. We feel that this is unlikely since polylysine is known to aggregate platelets but is not nearly as active as our derivative suggesting the existence of more specialized aggregating sites on collagen.21 Furthermore, polylysine does not initiate the platelet release reaction,22whereas glycine ethyl ester-collagen is normal in this regard (unpublished observations). Esterification of insoluble human collagen with methanol has been reported to fail to alter the potential to aggregate platelet^;^^,^^ however, a close scrutiny of the data indicates that it reduces the lag time for aggregation. The prevention of both polymerization and platelet aggregation by succinyl, acetyl, and trinitrobenzenesulfonyl collagens may well be due to the blocking of essential residues in the aggregating site or perhaps by the increase in negative charge of the molecule. Previous studies with acetyl, dinitrophenyl, and trinitrobenzenesulfonyl derivatives of insoluble adult human skin ~ o l l a g e n ~demonstrated 3,~~ that these collagens were no longer able to cause platelet aggregation, emphasizing the importance of polar groups t o the aggregation response. Also, succinylated collagen did not alter the ability t o aggregate platelets; this is in opposition to our observation that succinylated tropocollagen did not aggregate platelets. This difference may be due to the different collagens, but it is difficult t o reconcile a modification whereby the substitution of amino groups by various reagents prevents aggregation, but where a succinyl group has no effect unless the increased negative charge promotes aggregation. We have also found that succinylated collagens prevents fibroblast attachment and clone formation is tissue culture (unpublished results). The effects of oleyl and cetyl collagen are particularly of interest. These alkyl groups, when crosslinked to free collagen carboxyl groups, prevent platelet aggregation. The mechanism for this is clearly different from the effect of glycine ethyl ester groups since the former prevents aggregation and the latter stimulates. Since the alkyl crosslinking procedure derivatizes only about 25% of
370
KAY ET AL.
available glycine ethyl ester-titratable groups, further blocking of these groups as in the oleyl-glycine ethyl ester derivative partly restores the aggregation property of the molecule but not tho ability to form polymers. Figure 2 demonstrates the dramatic change in platelet morphology induced by a small quantity (5 Mg/ml) of collagen-glycine ethyl ester. Within 5 sec, the platelets begin to undergo characteristic shape changes and form pseudopodia-like structures which are very clear a t 15 sec; by 30 sec, sizeable secondary aggregates are formed. It is important to note that these changes are analogous to those observed with other aggregating agents and that no collagen fibers are visible. These platelet responses are of interest when considering the modification and design of collagen biomedical materials. I n our experience with collagenous (facia lata) heart valves, we found that hydrophobic surface coatings, such as olcyl or octadecyl coatings, prevented unwanted cellular proliferation and contributed to the stability of the p ro ~ th esis.~This hydrophobic effect would also now appear to be antithrombogenic, analogous perhaps to the antiadhesive properties of a l b ~ m i n ~coatings. ~ - ~ ~ One of the attractivc features of reacting collagens destined for biomaterial application with alkylamines is that the alkyl-peptide bond would conceivably be resistant to endogenous proteolytic attack since it would be a rather unfamiliar residue to most proteases. We suspect that such coatings would have good flow properties, but this remains to be tested.
Fig. 2. Changes in morphology of canine platelets in response to collagenglycine ethyl eater ( 5 pglrnl). Micrographs were taken with a Cambridge scanning electron microscope after fixation in 1.7% glutaraldehyde, dehydration in iaoamyl alcohol, critical point-drying, and gold plating: (a) 5 sec ( X 10,000); (b) 15 aec ( X 5000); (c) 30 sec ( X 10,000).
COLLAGEN DERIVATIVES AND PLATELET AGGREGATION 371
I n opposition to alkyl collagens, the use of collagen-glycine ethyl ester prostheses would be likely disasterous; however, since the promotion of tissue adhesiveness is the object of synthetic or nonit may well be that these synthetic wound or burn kinds of modifications would be conducive to rapid healing. This work was supported by grants from the Canadian Heart Foundation. The authors are indebted to Dr. F. Ghadially for the electron microscopy.
References 1. E . J. Miller and V. J. Matukas, Fed. Proc., 33, 1197 (1974). 2. A. S. Trimble and F. S. Metni, Can. Med. Assoc. J . , 105, 715 (1971). 3. B. Bharadwaj, E. Kurylo, and W. W. Kay, Ann. Thorac. Surg., 19, 158 (1975). 4. F. 0. Bowman, W. D. Hancock, and J. K. Malm, Arch. Surg., 107,724 (1974). 5. A. 1). Schwop, I). L. Wise, K. W. Sell, W. A. Skornick, and D. P. Dressler, Trans. Amer. Artif. Int. Organs, 20, 103 (1974). 6. S. J. Gourlay, R. M. Rice, A. F. Heggeli, J. C. Eaton, Jr., J. W. Hodge, Jr., and C. R. Wade. Trans. Amer. Artzf. Int. Organs, 21, 28 (1975). 7. N. J. Tavis, J. H. Harney, J. W. Thorriton, and R. H. Bartlett, J . Biomed. Mater. Res., 9, 285 (1975). 8. T. Nishihara, A. L. Rubin, and K. H. Stenzel, Trans. Amer. SOC.Artif. Znt. Organs, 13, 243, (1967). 9. A. L. Rubin, R. R . Riggio, R. L. Nachman, G. H. Schwartz, T. Miyata, and K. H. Stenzel, Trans. Amer. SOC.Artif. Znt. Organs, 14, 169 (1968). 10. K. H . Stenzel, J. F. Sullivan, T. Miyata, and A. L. Rubin, Trans. Amer. SOC.Art. Int. Organs, 15, 114 (1969). 11. D. F. Gibbons, Ann. Rev. Biophys. Bzoeng., 4, 367 (1975). 12. B. I). Ratner, T. Horbett, and A. S. Hoffman, J . Bzomed. Muter. Res., 9, 407 (1975). 13. A. Schwartz, K. H. Stenzel, I. Kohno, 1). Ast, T. Mioyata, and A. L. Rubin, J . Bzomed. Muter. Res., 9, 453 (1975). 14. J. Sevedenborg and S. I. Schwartz, J . Surg. Res., 19, 133 (1975). 15. R. L. Ehrman and G. P. Gey, J . Nut. Cancer Inst., 116, 1375 (1956). 16. G. V. It. Born, Nature (London), 194, 927 (1962). 17. K. A. I'iez and D. A. Torchia, Nature, 258, 87 (1975). 18. R. Muggli and H. R. Baumgartner, Thromb. Res., 3, 715 (1973). 19. R. Jaffe and 1). Deykin, J . Clin. Invest., 53, 875 (1974). 20. W. Schneider, W. Kubler, and R. Gross, Thromb. Diath. Haemorrh., 20, 314 (1968). 21. I).Puett, B. K. Wasserman, J . I). Ford and L. W. Cunningham, J . Clin. Invest., 52, 2495 (1973). 22. C. S. P. Jenkins, M. A. Packham, and I:. L. Kinlough-Rathbone, Blood, 37, 395 (1971). 23. G. D. Wilner, H . L. Nossel, and E. C. Leltoy. J . Ctin. Invest., 47,2616 (1968). 24. G. D. Wilner, H. I,. Nossel, and T. L. Procupez. Amer. J . Physiol., 220, 1074 (I97 1 ) .
372
KAY ET AL.
25. M. A. Packham, G. Evans, M. F. Glynn, and J. F . Mustard, J . Lab. Clin. Med., 73, 686 (1969). 26. P. J. Lyman, K. G. Klein, J. L. Brash, and B. K. Fritzinger, Thromb. Diath. Haemorrh. Suppl., 42, 109 (1970). 27. R. G. Mason, R. W. Shermer, and W. H. Zucker, Amer. J . Pathol., 73, 183 (1973). 28. D. Domurads, I). Thomas, and G. Braun, J . Biomed. Muter. Res., 9, 109 (1975).
Received February 9, 1976 Revised July 18, 1976
VOL. 11, PP. 365-372 (1977)
Inhibition and Enhancement of Platelet Aggregation by Collagen Derivatives W. W. KAY,* E. KURYLO, G. CHONG, and B. BHARADWAJ, Department of Surgery, University of Saskatchewan, Saskatoon, Saskatchewan, Canada, S7N OW0
Summary Rat tail tendon tropocollagen rapidly caused the aggregation of blood platelets coincident with collagen multimerization. Several long-chain alkyl derivatives of collagen] however, were completely unable to induce platelet aggregation even at unusually high concentrations. These collagen derivatives varied in their polymerization properties, and the polymerization of tropocollagen derivatives was shown not necessarily to be a prerequisite for platelet aggregation. These collagen derivatives may enhance the antithrombogenic properties of collagenbaaed prostheses. A short-chain derivative, collagen-glycine ethyl ester, profoundly enhanced platelet aggregation in the absence of any measurable polymerization. This derivative may be useful as a growth-promoting burn dressing.
INTRODUCTION Collagen is the most abundant and ubiquitous animal protein, largely functioning as supporting structures for various tissues. This unique protein is generally present in four genetically distinct types coded for by a t least five structural genes.' The evolution of such specialized proteins provides a strong, malleable, and chemically versatile molecule in abundance which is amendable to chemical modification and, as such, is a valuable material for the construction of biomedical prostheses. Indeed, various collagen polymers and copolymers are proving t o be highly advantageous alternatives t o more conventional medical pros these^,*-^ wound dressing^,^-^ and dialysis membranes. s-lo * Present address : Department of Bacteriology and Biochemistry] University of Victoria, Victoria, B.C. Canada VSW 2Y2. 365 @ 1977 by John Wiley & Sons, Inc.
366
KAY ET AL.
One of the major drawbacks to the use of various collagen prosthetic devises is the biological incompatability with respect to blood platelet adhesion. 11-15 Since we have recently reported that modified facia lata tricuspid valves of free-floating atrial grafts sustained unusually high longevity in dogs, we sought to assess the degree of antithrombogenicity of our grafts using the platelet-collagen aggregation response as an indicator of thrombogenic activity. To this end we described herein the preparation and properties of, as well as platelet responses to, various chemically modified tropocollagens which are either highly thrombogenic or antithrombogenic.
MATERIALS AND METHODS Rat tail tendon collagen was prepared by acetic acid extraction,15 exhaustive dialysis, and lyophilization. Glycine ethyl ester was coupled t o collagen carboxyl groups with a water-soluble carbodiimide [l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide]16 and the degree of modification estimated with 14C-glycineethyl ester (1.62 mmol/g collagen). Either oleylamine or cetylamine (66.5 mM) was crosslinked to collagen in 95y0 ethyl alcohol with dicyclohexylcarbodiimide (88.7 mM), and the degree of crosslinking estimated with 3H-octadecylamine (New England Nuclear, custom tritiation). Oleylamine-collagen-glycine ethyl ester was prepared after reaction with oleylamine by similar methods. Succinylated collagen was prepared by reacting collagen with 1 M succinic anhydride in 0.1 M NaHC03. Succinic anhydride was added stepwise a t 0°C over a period of 4 hr. After 12 hr, succinic anhydride was again added (0.5 M ) and was reacted for an additional 24 hr a t room temperature, dialyzcd, Iyophilized, and redisolved a t pH 8.0. Acetylated collagen was prepared by reacting collagen in saturated NaHCO, with 0.5 M acetic anhydride for 19 hr at 0°C. Trinotrobenzene collagen was prepared by reaction of rat tail collagen with trinotrobenzenesulfonic acid (1.66 mM) in 20/, acetic acid for 24 hr a t room temperature. All derivatives werp exhaustively dialyzed and Iyophilized prior to use. Blood was collected from the jugular vcin of a dog and mixed immediately with ACD (1:6 v/v). Platelet-rich plasma for aggregation studies was obtaincd after centrifugation a t 120 g’s for 20 min. Aggrpgation was measured on a T’eyton aggr:grcyyncltcressentially by the method of Born.16
COLLAGEN DERIVATIVES AND PLATELET AGGREGATION 367
Collagen polymerization was measured by reflected 420 nm light on a n Aminco microspectrophotometer operated as a nephelometer. Polymerization was followed at 37°C in 0.18 M Tris HCl (pH 7.3). Platelet-collagen interactions were followed directly by scanning electron microscopy on a Cambridge stereoscanning electron microscope. Platelets in plasma were fixed with 1.7% glutaraldehyde, caught on Nucleopore membrane filters, dehydrated in isoamyl alcohol, and critical point-dried. The samples were then plated with gold.
RESULTS AND DISCUSSION All collagen derivatives used in this study were primarily blocked a t free amino or carboxyl groups and, as predicted,” this should interfere with polymerization by disrupting charge group interactions. Clearly, these procedures largely inhibited collagen polymerization (Table I), with the exception of oleyl collagen which could still polymerize, albeit incompletely. Commensurate with the inhibition of polymerization, five of the seven collagen derivatives were subsequently unable to initiate platelet aggregation; however, both glycine ethyl ester and the oleyl-glycine ethyl ester derivatives were TABLE I Platelet Aggregation by Rat Tail Tropocollagen Derivatives
Collagen Derivative Native collagen Glyeine ethyl ester-collagen
Oleyl collagen Oleyl-glycine ethyl ester Cetyl/collagen Succinyl/collagen Acetyl/collagen Trinitrobenzenesulfonyl collagen
Concentration (rg/ml) 100 100 5.0 0.5 500 220 140 300 200 230
Aggregation Time. (set) 120 10 10
28 >20 hr 60 >20 hr >20 hr >20 hr >20 hr
Polymerization (relative intensity)b 100 0 0 0 48 0 15 12 0 12
* Defined as the time elapsed between the addition of collagen and the increase in light transmittance at the beginning of platelet aggregation. b Polymerization was estimated in relative intensity units with rat tail collagen at 100 pg/ml at 10 units. Maximum polymerization in 2 hr was recorded.
KAY ET AL.
368
still potent platelet aggregating agents, particularly glycine ethyl ester-collagen which exhibited a 200-fold increase in aggregating potency relative to native rat tail collagen. Collagen polymerization was found to occur in plasma as well under these conditions. Scanning electron microscopy readily showed extensive fiber formation. I n addition, 1251-collagenscan be readily sedimented by ultracentrifugation (unpublished results). Typical aggregation curves for some of these derivatives are shown in Figure 1. The immediate and complete aggregation by glycine ethyl ester-collagen is compared to the time-dependent aggregation of rat, tail tropocollagen. The
W
0
z
4 + !z
5
tn z 4
K
c W
> F
5W K
I-
1 min
Fig. 1. Aggregation of canine platelets by native collagen and modified collagens: (A) collagen-glycine ethyl ester (5 fig/ml) ; (B) oleyl-collagen-glycine ethyl ester (110 pg/ml); (C) native acid soluble rat tail tropocollagen (212 pg/ml); (D) Acyl-collagen (200 pg/ml). No changes were found with oleyl- or cetylcollagens.
COLLAGEN DERIVATIVES A N D PLATELET AGGREGATION 369
time dependence is thought by some investigators to represent p o l y m e r i ~ a t i o n , 'although ~ ~ ~ ~ from the results of Table I, the nonpolymerizing derivative collagen-glycine ethyl ester is an unusually potent aggregating agent by a t least an order of magnitude greater than any collagen that we are aware of. This may possibly be due to the creation of a basic collagen since the aid groups have been blocked. We feel that this is unlikely since polylysine is known to aggregate platelets but is not nearly as active as our derivative suggesting the existence of more specialized aggregating sites on collagen.21 Furthermore, polylysine does not initiate the platelet release reaction,22whereas glycine ethyl ester-collagen is normal in this regard (unpublished observations). Esterification of insoluble human collagen with methanol has been reported to fail to alter the potential to aggregate platelet^;^^,^^ however, a close scrutiny of the data indicates that it reduces the lag time for aggregation. The prevention of both polymerization and platelet aggregation by succinyl, acetyl, and trinitrobenzenesulfonyl collagens may well be due to the blocking of essential residues in the aggregating site or perhaps by the increase in negative charge of the molecule. Previous studies with acetyl, dinitrophenyl, and trinitrobenzenesulfonyl derivatives of insoluble adult human skin ~ o l l a g e n ~demonstrated 3,~~ that these collagens were no longer able to cause platelet aggregation, emphasizing the importance of polar groups t o the aggregation response. Also, succinylated collagen did not alter the ability t o aggregate platelets; this is in opposition to our observation that succinylated tropocollagen did not aggregate platelets. This difference may be due to the different collagens, but it is difficult t o reconcile a modification whereby the substitution of amino groups by various reagents prevents aggregation, but where a succinyl group has no effect unless the increased negative charge promotes aggregation. We have also found that succinylated collagens prevents fibroblast attachment and clone formation is tissue culture (unpublished results). The effects of oleyl and cetyl collagen are particularly of interest. These alkyl groups, when crosslinked to free collagen carboxyl groups, prevent platelet aggregation. The mechanism for this is clearly different from the effect of glycine ethyl ester groups since the former prevents aggregation and the latter stimulates. Since the alkyl crosslinking procedure derivatizes only about 25% of
370
KAY ET AL.
available glycine ethyl ester-titratable groups, further blocking of these groups as in the oleyl-glycine ethyl ester derivative partly restores the aggregation property of the molecule but not tho ability to form polymers. Figure 2 demonstrates the dramatic change in platelet morphology induced by a small quantity (5 Mg/ml) of collagen-glycine ethyl ester. Within 5 sec, the platelets begin to undergo characteristic shape changes and form pseudopodia-like structures which are very clear a t 15 sec; by 30 sec, sizeable secondary aggregates are formed. It is important to note that these changes are analogous to those observed with other aggregating agents and that no collagen fibers are visible. These platelet responses are of interest when considering the modification and design of collagen biomedical materials. I n our experience with collagenous (facia lata) heart valves, we found that hydrophobic surface coatings, such as olcyl or octadecyl coatings, prevented unwanted cellular proliferation and contributed to the stability of the p ro ~ th esis.~This hydrophobic effect would also now appear to be antithrombogenic, analogous perhaps to the antiadhesive properties of a l b ~ m i n ~coatings. ~ - ~ ~ One of the attractivc features of reacting collagens destined for biomaterial application with alkylamines is that the alkyl-peptide bond would conceivably be resistant to endogenous proteolytic attack since it would be a rather unfamiliar residue to most proteases. We suspect that such coatings would have good flow properties, but this remains to be tested.
Fig. 2. Changes in morphology of canine platelets in response to collagenglycine ethyl eater ( 5 pglrnl). Micrographs were taken with a Cambridge scanning electron microscope after fixation in 1.7% glutaraldehyde, dehydration in iaoamyl alcohol, critical point-drying, and gold plating: (a) 5 sec ( X 10,000); (b) 15 aec ( X 5000); (c) 30 sec ( X 10,000).
COLLAGEN DERIVATIVES AND PLATELET AGGREGATION 371
I n opposition to alkyl collagens, the use of collagen-glycine ethyl ester prostheses would be likely disasterous; however, since the promotion of tissue adhesiveness is the object of synthetic or nonit may well be that these synthetic wound or burn kinds of modifications would be conducive to rapid healing. This work was supported by grants from the Canadian Heart Foundation. The authors are indebted to Dr. F. Ghadially for the electron microscopy.
References 1. E . J. Miller and V. J. Matukas, Fed. Proc., 33, 1197 (1974). 2. A. S. Trimble and F. S. Metni, Can. Med. Assoc. J . , 105, 715 (1971). 3. B. Bharadwaj, E. Kurylo, and W. W. Kay, Ann. Thorac. Surg., 19, 158 (1975). 4. F. 0. Bowman, W. D. Hancock, and J. K. Malm, Arch. Surg., 107,724 (1974). 5. A. 1). Schwop, I). L. Wise, K. W. Sell, W. A. Skornick, and D. P. Dressler, Trans. Amer. Artif. Int. Organs, 20, 103 (1974). 6. S. J. Gourlay, R. M. Rice, A. F. Heggeli, J. C. Eaton, Jr., J. W. Hodge, Jr., and C. R. Wade. Trans. Amer. Artzf. Int. Organs, 21, 28 (1975). 7. N. J. Tavis, J. H. Harney, J. W. Thorriton, and R. H. Bartlett, J . Biomed. Mater. Res., 9, 285 (1975). 8. T. Nishihara, A. L. Rubin, and K. H. Stenzel, Trans. Amer. SOC.Artif. Znt. Organs, 13, 243, (1967). 9. A. L. Rubin, R. R . Riggio, R. L. Nachman, G. H. Schwartz, T. Miyata, and K. H. Stenzel, Trans. Amer. SOC.Artif. Znt. Organs, 14, 169 (1968). 10. K. H . Stenzel, J. F. Sullivan, T. Miyata, and A. L. Rubin, Trans. Amer. SOC.Art. Int. Organs, 15, 114 (1969). 11. D. F. Gibbons, Ann. Rev. Biophys. Bzoeng., 4, 367 (1975). 12. B. I). Ratner, T. Horbett, and A. S. Hoffman, J . Bzomed. Muter. Res., 9, 407 (1975). 13. A. Schwartz, K. H. Stenzel, I. Kohno, 1). Ast, T. Mioyata, and A. L. Rubin, J . Bzomed. Muter. Res., 9, 453 (1975). 14. J. Sevedenborg and S. I. Schwartz, J . Surg. Res., 19, 133 (1975). 15. R. L. Ehrman and G. P. Gey, J . Nut. Cancer Inst., 116, 1375 (1956). 16. G. V. It. Born, Nature (London), 194, 927 (1962). 17. K. A. I'iez and D. A. Torchia, Nature, 258, 87 (1975). 18. R. Muggli and H. R. Baumgartner, Thromb. Res., 3, 715 (1973). 19. R. Jaffe and 1). Deykin, J . Clin. Invest., 53, 875 (1974). 20. W. Schneider, W. Kubler, and R. Gross, Thromb. Diath. Haemorrh., 20, 314 (1968). 21. I).Puett, B. K. Wasserman, J . I). Ford and L. W. Cunningham, J . Clin. Invest., 52, 2495 (1973). 22. C. S. P. Jenkins, M. A. Packham, and I:. L. Kinlough-Rathbone, Blood, 37, 395 (1971). 23. G. D. Wilner, H . L. Nossel, and E. C. Leltoy. J . Ctin. Invest., 47,2616 (1968). 24. G. D. Wilner, H. I,. Nossel, and T. L. Procupez. Amer. J . Physiol., 220, 1074 (I97 1 ) .
372
KAY ET AL.
25. M. A. Packham, G. Evans, M. F. Glynn, and J. F . Mustard, J . Lab. Clin. Med., 73, 686 (1969). 26. P. J. Lyman, K. G. Klein, J. L. Brash, and B. K. Fritzinger, Thromb. Diath. Haemorrh. Suppl., 42, 109 (1970). 27. R. G. Mason, R. W. Shermer, and W. H. Zucker, Amer. J . Pathol., 73, 183 (1973). 28. D. Domurads, I). Thomas, and G. Braun, J . Biomed. Muter. Res., 9, 109 (1975).
Received February 9, 1976 Revised July 18, 1976
