Fibrinogen-Conjugated Albumin Polymers and Their Interaction with

and Department of Internal Medicine and Blood Center, School of Medicine, ... DaShawn A. Hickman , Christa L. Pawlowski , Ujjal D. S. Sekhon , Joy...
0 downloads 0 Views 81KB Size
Download PDF
Biomacromolecules 2001, 2, 1192-1197

1192

Fibrinogen-Conjugated Albumin Polymers and Their Interaction with Platelets under Flow Conditions Shinji Takeoka,† Yuji Teramura,† Yosuke Okamura,† Makoto Handa,‡ Yasuo Ikeda,‡ and Eishun Tsuchida*,† Department of Polymer Chemistry, Advanced Research Institute for Science and Engineering, Waseda University, Tokyo 169-8555, Japan; and Department of Internal Medicine and Blood Center, School of Medicine, Keio University, Tokyo, 160-8582, Japan Received May 1, 2001; Revised Manuscript Received August 9, 2001

Albumin polymers, having an average diameter of 1020 ( 250 nm, were prepared by the disulfide polymerization of recombinant human serum albumin (rHSA) by controlling of the pH and temperature. Fibrinogen could be conjugated on the surface of an albumin polymer using N-succinimidyl 3-(2pyridyldithio)propionate (SPDP). Under flow conditions, the fibrinogen-conjugated albumin polymers (fibrinogen-albumin polymers) were irreversibly attached to the platelet-immobilized surface in the reconstituted blood at a low platelet concentration ([platelet] ) 5.0 × 104/µL, a 5-fold diluted platelet concentration), and the attachment was suppressed by the addition of anti-GPIIb/IIIa monoclonal antibodies. It was confirmed that fibrinogen-albumin polymers specifically interacted with GPIIb/IIIa expressed on the surface of the activated platelets. Although platelets with a low platelet concentration were hardly attached to the platelet-immobilized surface under the flow conditions, the addition of fibrinogen-albumin polymers enhanced the attachment of the remaining platelets to the surface, indicating that the fibrinogen-albumin polymers would help the hemostatic ability of platelets at the site of vascular injury of patients in thrombocytopenia. Introduction Platelet transfusion plays an important role as a supportive therapy for the treatment of cancer or hematologic malignancies or during surgical procedures. Recently, the number of applications of platelet concentrates is increasing even though the shortage of platelets has continued due to the short-term storage of platelet concentrates (3 days in Japan). Therefore, it is difficult for stored platelets to be available for use in emergency care such as natural disasters. Another problem is the risk of viral and bacterial infections during transfusions. The development of platelet substitutes is necessary in order to solve these issues. Platelet substitutes, which have so far been developed, utilize the recombinant fragment of the platelet membrane such as glycoprotein (GP) IbR1,2 and the GPIa/IIa complex3 or fibrinogen,4,5 RGDS,6 etc., as the recognition sites. Recently, Levi et al succeeded in reducing the bleeding time with fibrinogen-coated albumin microcapsules.5 Albumin-based particles such as albumin microcapsules, nanospheres, or microspheres have been used as intravenously injectable particles since the 1960s because of their high biocompatibility and biodegradability.7-9 Since the preparation methods of albumin-based particles include spray-drying, coacervation, and emusification using organic solvents, detergents, or cross-linkers as stabilizers, it was * Corresponding author. E-mail: [email protected]. Telephone: +81-5-3286-3120. Fax: +81-3-3205-4740. † Waseda University. ‡ Keio University.

difficult to completely remove these additives and control the particle diameter.8 We have already succeeded in the preparation of albumin polymers, which were polymerized with intermolecular disulfide bonds by changing the pH and temperature in an aqueous system and found that the particle diameter could be controlled from 50 to 300 nm.10 It was confirmed that the addition of ristocetin to the rGPIbRconjugated albumin polymers (rGPIbR-albumin polymers) having a 240 nm diameter in the presence of the von Willebrand factor (vWf) caused specific aggregation involving the remaining platelets, and the rGPIbR-albumin polymers also became specifically attached to the vWfimmobilized surface under flow conditions.11 In this study, for the purpose of constructing the particle system which recognizes the platelets attached to the site of vascular injury, we selected fibrinogen, which specifically recognizes GPIIb/IIIa expressed on the activated platelet surface. Fibrinogen is a 340 kDa fibrious glycoprotein having two identical disulfide-linked subunits composed of three nonidentical polypeptide chains: AR, Bβ, and γ present at 0.3 g/dL in plasma, which plays an important role in a blood coagulation system. During bleeding, this protein plays a critical role in secondary hemostasis because of being involved in platelet aggregation.12,13 As mentioned above, Levi et al. used fibrinogen-coated albumin microcapsules as platelet substitutes, of which the surface fibrinogen was physically adsorbed. On the other hand, we covalently conjugated fibrinogen to the surface of the albumin polymers with N-succinimidyl 3-(2-pyridyldithio)propionate (SPDP).14

10.1021/bm015554o CCC: $20.00 © 2001 American Chemical Society Published on Web 11/02/2001

Fibrinogen-Albumin Polymers

Biomacromolecules, Vol. 2, No. 4, 2001 1193

Scheme 1. Synthesis of Albumin Polymers

Scheme 2. Conjugation of Fibrinogen to the Albumin Polymers

Our first objective is that fibrinogen-conjugated albumin polymers would be confirmed to be effective as Levi’s substitutes. Moreover, to basically and quantitatively analyze in vitro, we investigated the interaction of the fibrinogenalbumin polymers with GPIIb/IIIa expressed on the activated platelet surface and the recruiting of circulating platelets. Materials and Methods Synthesis of the Albumin Polymers (Scheme 1). The albumin polymers, of which the average diameter is 1-3 µm, was based on the preparation method of albumin polymers of which the average diameter is 240 nm.10 A recombinant human serum albumin (rHSA, 25 g/dL) was kindly donated by Welfide Co. (Osaka, Japan) and dialyzed against pure water for 12 h at 4 °C to remove the stabilizers such as N-acetyl-D,L-tryptophan and sodium caprate. After dilution with saline to 1.0 g/dL, 800 µL of 0.1N NaOH solution was added to the rHSA solution (25 mL) until the pH became 10.65 at room temperature. After being heated at 80 °C for 10 min, the solution was cooled in an ice bath for 10 min. The solution was stirred at room temperature for 10 min, the pH became 10.51. Then, 1 mL of 0.1N HCl solution was added until the pH of the solution became 5.90, and a white and transparent rHSA solution was obtained. During stirring the solution for 90 min at 40 °C, the solution gradually became turbid. After the addition of excess iodoacetamide (25 mg) as a terminant of polymerization at room temperature, the solution was dialyzed against the phosphate-buffered saline (PBS) for 20 h at 5 °C, and the 25 mL dispersion of albumin polymers ([HSA] ) 0.9 g/dL, pH 7.4) were prepared. The average diameter was determined by a dynamic scattering method (Coulter particle analyzer, model N4SD, Coulter Co., Fullerton, CA) and scanning electron microscopy (SEM, Hitachi Ltd., Tokyo). Conjugation of Fibrinogen to the Albumin Polymers (Scheme 2). The method to conjugate fibrinogen to the surface of albumin polymers with N-succinimidyl 3-(2pyridyldithio)propionate (SPDP, Wako Pure Chemical Co., Osaka) was followed by our previous method.8 To 15 mL of the albumin polymers dispersion (1.6 g/dL) was added 60 µL of a 5 mM SPDP ethanol solution, and this was stirred

for 30 min at room temperature. After the precipitation by centrifugation (6000 rpm, 5 min) and the redispersion of the albumin polymers, 12 mL of pyridyldithio-bound albumin polymers (PD-albumin polymers, [HSA] ) 1.7 g/dL) was obtained. Human fibrinogen (Calbiochem, San Diego, CA) was dissolved into PBS (pH7.4), and the concentration was adjusted to 1.0 g/dL. To the solution (500 µL) was added a 5 µL aliquot of 5 mM SPDP ethanol solution and incubated for 20 min at room temperature. After the addition of 10 µL of a 1 M dithiothreitol (DTT) solution at room temperature, 800 µL of SH-bound fibrinogen (SH-fibrinogen, [fibrinogen] ) 0.66 g/dL) was obtained by the separation from unreacted SPDP, DTT, and byproducts with gel permeation chromatography (15 mm o.d. × 55 mm h, Sephadex G25, Pharmacia Co.). A 600 µL of SH-fibrinogen (0.66 g/dL) was mixed with the 5 mL of PD-albumin polymers dispersion ([HSA] ) 1.7 g/dL) at 25 °C for 12 h, and fibrinogen-conjugated albumin polymers (fibrinogen-albumin polymers) were prepared by the removal of unreacted fibrinogen by centrifugation (4000 rpm, 5 min). The concentration of fibrinogen conjugated to the surface of the albumin polymers was determined with a sandwich enzymelinked immunosorbent assay (ELISA) with using two kinds of antibody; monoclonal anti-human fibrinogen (Dako) and polyclonal peroxidase-conjugated rabbit anti-human fibrinogen (Dako). Preparation of Reconstituted Blood. Blood drawn from healthy volunteers was mixed with a 10:1 volume of acidcitrate-dextrose composed of 2.2% (w/v) sodium citrate, 0.8% (w/v) citric acid, and 2.2% (w/v) glucose (ACD).2 The blood was centrifuged at 600 rpm for 15 min at room temperature, and the platelet-rich plasma (PRP) was replaced with an equal volume of a 0.9% NaCl solution containing 10% (v/v) ACD (10% ACD-saline). After red blood cells were resuspended and centrifuged at 2500 rpm for 10 min at room temperature, the supernatant was replaced with 10% ACD-saline. Each procedure was repeated twice, and the buffy coat, the layer of white blood cells, was completely removed. For perfusion studies, the red blood cells were reconstituted to 40% of hematocrit (Hct) in a Hepes-Tyrode buffer (H-T buffer). The residual platelet concentration was

1194

Biomacromolecules, Vol. 2, No. 4, 2001

Figure 1. Turbidity change of the albumin solution at 40 °C.

(1.20 ( 0.2) × 104/µL. The Hct and platelet concentration were determined with an automated hematology analyzer (K4500, SYSMEX, Kobe, Japan). Preparation of Collagen-Immobilized Surface. Collagen I-A (3.0 mg/mL, Cellmatrix, Nitta Gelatin Inc., Osaka, Japan) was dispersed in PBS (pH7.3) at 4 °C to give a final concentration of 30 µg/mL. A glass plate (diameter, 24 mm; thickness, 0.5 mm) was immersed into the collagen dispersion at 4 °C for 8 h. Preparation of Platelet-Immobilized Surface. Blood drawn from healthy volunteers was mixed with a 9:1 volume of ACD. The blood was centrifuged at 700 rpm for 15 min at room temperature, and then PRP was obtained. The PRP was mixed with a 3:20 volume of ACD containing 1 µM prostagrandin E1 (PGE1, Sigma, St. Louis, MO). After the PRP-ACD solution was centrifuged at 2500 rpm for 7 min at room temperature, the plasma was replaced with a Ringer’s-citrate-dextrose solution (RCD solution, 0.15 g/dL citric acid, 0.018 g/dL glucose, 0.0085 g/dL MgCl2, 0.0076 g/dL KCl, 0.12 g/dL NaCl, pH 6.5) containing 1 µM PGE1. After the platelets resuspended in the RCD solution were centrifuged at 2500 rpm for 7 min at room temperature, platelets were resuspended to be 2.0 × 105/µL using an H-T buffer. Into this platelet suspension, a collagen-immobilized glass plate was immersed at 37 °C for 1 h and rinsed carefully with PBS before used for perfusion studies. Measurement of the Interaction of Fibrinogen-Conjugated Albumin Polymers with the Platelet-Immobilized Surface. The interaction of fluorescein isothiocyanate (FITC)or rhodamine-labeled fibrinogen-albumin polymers with the platelet-immobilized surface was observed using a recirculating chamber mounted on an epifluorescence microscopy (ECLIPS TE300, Nikon, Tokyo) equipped with a CCD camera.15 All the perfusion studies were performed at 37 °C. Results and Discussion We modified the preparation method of the albumin polymers of which the diameter was submicrometer as previously reported.10 After heating a rHSA solution (pH10.65, 25 mL) at 80 °C for 10 min and then mixing with 1 mL of 0.1N HCl solution, a white transparent albumin solution (pH5.9) was obtained. The solution turbidity (∆OD) was increased during stirring at 40 °C as shown in Figure 1. As

Takeoka et al.

the turbidity of the rHSA solution increased, the particles of the albumin polymers grew from submicrometers to micrometers. For example, when the ∆OD of the albumin solution became 0.4 ( 0.1, the particle diameter of the supernatant after being centrifuged at 12 000 rpm for 5 min was 310 ( 50 nm (yield 95%). When the ∆OD became 1.8 ( 0.2, the albumin polymers, of which diameter was from several hundred nanometers to several micrometers, were obtained. The albumin polymers with various diameters could be separated by changing the centrifugation conditions. The dispersion was centrifuged at 2000 rpm for 5 min to obtain a supernatant. The particle diameter of the precipitates after being centrifuged at 6000 rpm for 5 min from the supernatant was 1800 ( 200 nm (yield 40%). We used the thus obtained albumin polymers in the following experiments. Human serum albumin (HSA) has 17 pairs of disulfide bonds and one thiol group, and its molecular weight is 66.5 kDa.16 These disulfide bonds and the thiol group, which usually exist inside the HSA molecule under physiological conditions, should be exposed to the aqueous phase after conformational conversion from the N-form to the B- or A-form under alkaline conditions (pH10.65). The thioldisulfide exchange reaction among one mercapto group at 34Cys and part of the disulfide bonds extensively occurs at the intra- and intermolecular levels at 80 °C, resulting in the formation of albumin oligomers.17,18 By lowering the solution pH near the isoelectric point (HSA; pI ) 4.9), the electrostatic repulsion decreases among the negatively charged albumin oligomers, resulting in the propagation of the polymerization. The particle diameter is controlled by the pH, temperature and reaction time. For the preparation method of the albumin polymers with 240 nm diameter, the reaction time at 80 °C, pH 10.65, was 20 min. On the other hand, the reaction time was shortened to 10 min for the preparation of larger albumin polymers because the degree of denaturation should be lower for controlling the degree of polymerization at 40 °C more easily. After the PD group was introduced into the fibrinogen by the reaction with SPDP, the introduction ratio of the PD group was determined by measuring the 2-thiopyridone (2TP) liberated from the PD group by reduction with DTT. The 2TP was separated and quantified by size-exclusion HPLC using a calibration curve for 2TP. The result that 1.3 PD groups were found by calculation to be introduced into the one fibrinogen indicates that the minimized modification could be performed. Thus SH-fibrinogen was simultaneously obtained after the separation of the PD with Sephadex G25. In the same way, a 0.6 PD group was found to be introduced into the one HSA covering the surface of the albumin polymers. The dispersion of PD-albumin polymers was then mixed with SH-fibrinogen at 25 °C for 10 h to prepare the fibrinogen-albumin polymers with a thiol-disulfide exchange reaction. From the ELISA measurement, about 3000 fibrinogen molecules were found by calculation to be conjugated to the surface of one albumin polymer. We expected that less than 10 000 fibrinogens were conjugated on the surface of one albumin polymer particle having a 1 µm diameter, because we had already confirmed that there were several tens of thousands of the other stable

Fibrinogen-Albumin Polymers

Figure 2. SEM image of albumin polymers.

Figure 3. Relative quantity of fibrinogen and fibrinogen conjugated to the surface of albumin polymers detected by an ELISA method in PBS (pH7.4) incubated at 4 and 25 °C. Fibrinogen: (b) 4 °C; (2) 25 °C. Fibrinogen-albumin polymer: (O) 4 °C; (4) 25 °C.

proteins such as GPIbR (45 kDa)19 and hemoglobin (65 kDa) conjugated on the same size of albumin polymer, and we considered the fibrous fibrinogen with molecular weight of 340 kDa. Fibrous shape shows the more steric hindrance, which would reduce the conjugation number in comparison with spherical compact shape. The second reason should be the denaturation of fibrinogen structure. Figure 3 shows the quantitative change of fibrinogen detected by an ELISA method. The starting point in the case of free fibrinogen is when the asreceived fibrinogen powder was dissolved in the solution, whereas, the point in the case of the fibrinogen-albumin polymers is when the frozen albumin polymers after conjugation with fibrinogen was thawed in the solution. The decay curves clearly indicate that the recognition site of the fibrinogen to the antibodies for the ELISA method should be denatured by the first-order kinetics. The conjugation of the fibrinogen to the albumin polymers did not influence on the kinetics as shown in Figure 3. This denies the possibility of the remaining enzymes such as plasminogen for decomposition of the fibrinogen, because the fibrinogen-albumin polymers should be separated from the enzymes by a size exclusion chromatography. Furthermore, even if such enzymes were also conjugated to the albumin polymers, the denaturation rate should become extensively low in comparison with the free fibrinogen due to the low collision frequency. We can also confirm the denaturation of the fibrinogen from the following phenomenon. In the ADP-induced platelet aggregation measured by an aggregometer, the aggregatory activity was confirmed by the addition of fibrinogen which was used immediately after the dissolution of the asreceived

Biomacromolecules, Vol. 2, No. 4, 2001 1195

frozen powder in the PRP solution. However, if the fibrinogen was previously dissolved in the buffer solution and stood at 25 °C for 20 h, the aggregatory activity was reduced to be about half. This indicates the denaturation of the recognition site of the fibrinogen to the GPIIb/IIIa of the activated platelet. From the decay curve in Figure 3, 60 or 70% of fibrinogen on an albumin polymer had already been denatured in normal handling. Therefore, a conjugation number of 3000 molecules would be reasonable. To minimize the influence of the denaturation, fibrinogen-albumin polymers should be frozen just after preparation and thawed before use. We have also confirmed that the aggregatory ability of the fibrinogen-albumin polymers was maintained even after the freeze-thawing procedure. Therefore, we can repeat the flowing experiment under the same conditions after thawing the frozen sample. To study the interaction between the fibrinogen-albumin polymers and activated platelets, the platelet-immobilized surface was used under flow conditions. The perfusion study was performed at 4 or 6 h after the preparation of the fibrinogen-albumin polymers in consideration of the denaturation of the fibrinogen. The platelets attached to the collagen surface are generally activated to express GPIIb/ IIIa on the platelet surface. First, the attachment of mepacrine-labeled platelets to the platelet-immobilized surface by circulating human whole blood was observed with a fluorescence microscope. The circulating platelets at a low shear rate such as 350 s-1 were immediately attached to the platelet-immobilized surface. This attachment was suppressed by the addition of an anti-GPIIb/IIIa monoclonal antibody, YM207, indicating the involvement of RGD-bearing ligands such as fibrinogen and von Willebrand factor in the whole blood and the expression of GPIIb/IIIa on the surface of the activated platelets. Second, for the fibrinogen-albumin polymers, their attachment to the platelet-immobilized surface was observed at low shear rates such as 350 s-1, where fibrinogen and GPIIb/IIIa interacted with each other. As shown in Figure 4, under flow conditions, the rhodamine-labeled fibrinogenalbumin polymers ([HSA] ) 0.02 g/dL, [fibrinogen] ) 0.06µg/dL) immediately attached to the surface in the presence of 1 mM calcium ion, whereas the rhodaminelabeled albumin polymers hardly attached to the plateletimmobilized surface. The attachment of the fibrinogenalbumin polymers was also suppressed by the addition of the anti-GPIIb/IIIa monoclonal antibody. As shown in Figure 5, in the absence of calcium ion, the fibrinogen-albumin polymers did not become attached to the platelet-immobilized surface. It was reported that the activation of GPIIb/IIIa by complexation with calcium ion was followed by the conformational change to bind fibrinogen.20 Therefore, GPIIb/ IIIa cannot bind fibrinogen in the absence of calcium ion. Since GPIIb/IIIa is a kind of integrins which mediates the cell-extracellular matrix interaction in a calcium ion-dependent manner,20,21 the attachment of the fibrinogen-albumin polymers to the platelet-immobilized surface was confirmed to be specific to the activated GPIIb/IIIa. It was also confirmed that the attached number of fibrinogen-albumin polymers decreased with the increasing shear rate as shown

1196

Biomacromolecules, Vol. 2, No. 4, 2001

Takeoka et al.

Figure 4. Observation of the rhodamine-labeled fibrinogen-albumin polymers attached to the platelet-immobilized surface with fluorescence microscopy.

Figure 5. Number of fibrinogen-albumin polymers attached to the platelet-immobilized surface under the flow conditions in the presence or absence of calcium ion.

Figure 6. Number of the fibrinogen-albumin polymers attached to the platelet-immobilized surface under the various shear rates.

in Figure 6. When the platelet concentration in the circulating reconstituted blood was 5.0 × 104/µL (5-fold diluted platelet concentration), the remaining platelets tended not to become attached to the platelet surface. The attached number of albumin polymers increased as the concentration of the fibrinogen-albumin polymers added to the blood at a low platelet concentration increased. Simultaneously, the attached number of remaining platelets also increased on the platelet surface as shown in Figure 7. The attached number of platelets was increased with the increasing number of the fibrinogen-albumin polymers attached to the platelet-

Figure 7. Effect on the attachment of remaining platelets to the platelet-immobilized surface by the addition of fibrinogen-albumin polymers. The concentrations of fibrinogen-albumin polymers are 0, 1.0 × 10-2, 2.1 × 10-2, 3.6 × 10-2, and 4.7 × 10-2 g/dL from the left plot, respectively.

immobilized surface, indicating that the flowing platelets were recruited by the fibrinogen-albumin polymers attached to the platelet surface. The important functions of microparticles as platelet substitutes are as follows: (1) recognition and attachment to the collagen exposed at the site of vascular injury, and (2) activation and recruit of the remaining and flowing platelets. We previously succeeded in the preparation of rGPIbR-conjugated albumin polymers, which showed vWfmediated platelet adhesion at high shear rates, and observed the specific attachment of the rGPIbR-albumin polymers to the vWf surface under flow conditions. In this study, the system using the platelet-immobilized surface is a model of the site of vascular injury, and we could quantitatively discuss the attachment of both the fibrinogen-albumin polymers and the remaining platelets to the activated platelets which adhered to the collagen. Levi et al. reported similar results such that fibrinogen-coated albumin microcapsules reduced bleeding time in thrombocytopenic rabbits and indicated that fibrinogen-based platelet substitutes were effective in reducing the bleeding time. However, we recognized that fibrinogen was unstable in solution. To maintain the activity of fibrinogen, we propose that fibrinogen-albumin polymers should be frozen just after preparation and stored before use. It is confirmed that the activity of fibrinogen-albumin polymer did not change after freeze-thawing. Furthermore, human-derived fibrinogen still has a risk of infection.

Fibrinogen-Albumin Polymers

Therefore, the development of platelet substitutes derived from completely artificial products should be the next target. For this purpose, we are trying to conjugate the RGDS sequence,22 which is a recognition peptide for GPIIb/IIIa, or the γC dodecapeptide sequence of fibrinogen, to the surface of the albumin polymers.23 Acknowledgment. The authors would thank Dr. T. Nishiya and Dr. K. Yokoyama at Keio University for useful discussions about the functional evaluation of fibrinogenalbumin polymers and Welfide Co. for donating rHSA, and Dr. H. Ohkawa at Waseda University, ARISE, for his advice in the preparation of albumin polymers. This work was supported by Health Science Research Grants (Artificial Platelets), Research on Advanced Medical Technology, Ministry of Health, Labor and Welfare, Japan, and grantsin-aid from the Ministry of Education, Culture, Sport, Science, and Technology, Japan (No. 11877172). References and Notes (1) Kitaguchi, T.; Murata, M.; Iijima, K.; Kamide, K.; Imagawa, T.; Ikeda, Y. Characterization of liposomes carrying von Willebrand factor-binding domain of platelet glycoprotein IbR: A potential substitute for platelet transfusion. Biochem. Biophys. Res. Commun. 1999, 261, 784-789. (2) Nishiya, T.; Murata, M.; Handa, M.; Ikeda, Y. Targetting of liposomes carrying recombinant fragments of platelet membrane glycoprotein IbR to immobilized von Willebrand factor under flow conditions. Biochem. Biophys. Res. Commun. 2000, 270, 755-760. (3) Nishiya T. Liposome-based platelet substitutes. Artif. Blood 2000, 8, 6-12. (4) Agam, G.; Livne, A. A. Erythrocytes with covalently bound fibrinogen as a cellular replacement for the treatment of thrombocytopenia. Eur J. Clin. InVest. 1992, 22, 105-112. (5) Levi, M.; Friederich, P. W.; Middleton, S.; De Groot, P. G.; Wu Y. P.; Harris, R.; Biemond, B. J.; Heijien, F. G.; Levin, J.; Ten Cate, J. W. Fibrinogen-coated albumin microcapsules reduce bleeding in severely thrombocytopenic rabbits. Nat. Med. 1999, 5, 107-111. (6) Coller, B. S.; Springer, K. T.; Beer, J. H.; Mohandas, N.; Scudder, L. E.; Norton, K. J.; West, S. M. Thromboerythrocytes. In vitro studies of a potential autologous, semi-artificial alternative to platelet transfusions. J. Clin. InVest. 1992, 89, 546-555. (7) Kramer, P. A. Albumin microspheres as vehicles for achieving specificity in drug delivery. J. Pharm. Sci. 1974, 63, 1646-1647. (8) Gupta, P. K.; Hung, C. T. Albumin microsphere I: physicochemical characteristics. J. Microencapsulation 1989, 6, 427-462.

Biomacromolecules, Vol. 2, No. 4, 2001 1197 (9) Burger, J. J.; Tomlinson, E.; Deroo, J. E.; Palmer, J. Technetium99m labering of albumin microspheres intended for drug targeting. Methods Enzymol. 1985, 112, 43-56. (10) Takeoka, S.; Teramura, Y.; Ohkawa, H.; Ikeda, Y.; Tsuchida, E. Conjugation of von Willebrand factor-binding domain of platelet glycoprotein IbR to size-controlled albumin microspheres. Biomacromolecules 2000, 1, 290-295. (11) Teramura, Y.; Takeoka, S.; Tsuchida, E.; Ikeda, Y. Preparation of size-controlled albumin polymers and evaluation of the GPIbR conjugates. Artif. Blood 2000, 8, 90-95. (12) Ikeda, Y.; Handa, M.; Kawano, K.; Kamata, T.; Murata, M.; Araki, Y.; Ando, H.; Kawai, Y.; Watanabe, K.; Itagaki, I.; Sakai, K.; Ruggeri, Z. M. The role of von Willebrand factor and fibrinogen in platelet aggregation under varying shear stress. J. Clin. InVest. 1991, 87, 1234-1240. (13) Kulkarni, S.; Dopheide, S. M.; Yap, C. L.; Ravanat, C.; Freund M.; Mangin, P.; Heel, K. A.; Street, A.; Harper, I. S.; Lanza, F.; Jackson, S. P. A revised model of platelet aggregation. J. Clin. InVest. 2000, 105, 783-791. (14) Carlsson, J.; Drevin, H.; Alex, R. Protein Thiolation and Reversible Protein-Protein conjugation. N-Succinimidyl 3-(2-pyridyldithio)propionate, a new heteronifunctional reagent. Biochem. J. 1978, 173, 723-737. (15) Kawakami, K.; Harada, Y.; Sakasita, M.; Nagai, H.; Handa, M.; Ikeda, Y. A new method for continuous measurement of platelet adhesion under flow conditions. ASAIO J. 1993, 39, M558-M560. (16) Peters, T., Jr. All about Albumin: biochemistry, genetics, and medical applications; Academic Press: San Diego, CA, 1996, pp 9-23. (17) Huggins, C.; Tapley, D. P.; Jensen, E. J. Sulphydryl-disulphide relationships in the induction of gels in proteins by urea. Nature 1951, 167, 592-593. (18) Warner, C. R.; Levy, M. Denaruration of bovine Plasma albumin. II. Isolation of intermediates and mechanism of the reaction at pH 7. J. Am. Chem. Soc. 1958, 80, 5735-5744. (19) Murata, M.; Ware, J.; Ruggeri, Z. M. Site-directed mutagenesis of a soluble recombinant fragment of platelet glycoprotein Iba demonstrating negatively charged residues involved in von Willebrand factor binding. J. Biol. Chem. 1991, 266, 15474-15480. (20) Phillips, D. R.; Charo, I. F.; Scarborough, R. M. GPIIb-IIIa: The Responsive Integrin. Cell 1991, 65, 359-362. (21) Phillips, D. R.; Charo, I. F.; Parise, L. V.; Fitzgerald, L. A. The platelet membrane glycoprotein IIb/IIIa complex. Blood 1988, 71, 831-843. (22) Ruoslahti, E.; Pierschbacher, M. D. Arg-Gly-Asp: a versatile cell recognition signal. Cell 1986, 44, 517-518. (23) Yokoyama, K.; Erickson, H. P.; Ikeda, Y.; Takada, Y. Identification of amino acid sequences in fibrinogen γ-chain and tenascin C C-terminal domains critical for binding to integrin Rvβ3. J. Biol. Chem. 2000, 275, 16891-16898.

BM015554O