Interactions of Apolipoproteins AI, AII, B and HDL, LDL, VLDL with Polyurethane and Polyurethane-PEO Surfaces.

The lipoproteins (HDL, LDL, VLDL) are important components of blood present in high concentration. Surprisingly, their role in blood-biomaterial interactions has been largely ignored. In previous work apolipoprotein AI (the main protein component of HDL) was identified as a major constituent of protein layers adsorbed from plasma to biomaterials having a wide range of surface properties, and quantitative data on the adsorption of apo AI to a biomedical grade polyurethane were reported. In the present communication quantitative data on the adsorption of apo AI, apo AII and apoB (the latter being a constituent of LDL and VLDL), as well as the lipoprotein particles themselves (HDL, LDL, VLDL), to a biomedical segmented polyurethane (PU) with and without an additive containing poly(ethylene oxide) (material referred to as PEO) are reported. Using radiolabeled apo AI, apo AII, and apoB, adsorption levels on PU from buffer at a protein concentration of 50 µg/mL were found to be 0.34, 0.40, and 0.14 µg/cm(2) (12, 23, and 0.25 nmol/cm(2)) respectively. Adsorption to the PEO surface was <0.02 µg/cm(2) for all three apolipoproteins demonstrating the strong protein resistance of this material. In contrast to the apolipoproteins, significant amounts of the lipoproteins were found to adsorb to the PEO as well as to the PU surface. X-ray photoelectron spectra, following exposure of the surfaces to the lipoproteins, showed a strong phosphorus signal, confirming that adsorption had occurred. It therefore appears that a PEO-containing surface that is resistant to apolipoproteins may be less resistant to the corresponding lipoproteins.

Surface modifications of Nitinol for biomedical applications.

Cathodic electrophoretic deposition (EPD) has been utilized for the fabrication of composite films for the surface modification of NiTi shape memory alloys (Nitinol). In the proposed method, chitosan (CH) was used as a matrix for the incorporation of other functional materials, such as heparin, hydroxyapatite and bioglass. Chitosan-heparin films were deposited from solutions of non-stoichiometric chitosan-heparin complexes. It was found that the addition of anionic heparin to the solutions of cationic chitosan resulted in a significant increase in the cathodic deposition rate. The thickness of the films prepared by this method varied in the range of 0.1-3 microm. The ability of the chitosan-heparin films to bind antithrombin, as measured by binding of (125)I-radiolabeled antithrombin, was much greater than that of pure chitosan films. Composite chitosan-hydroxyapatite films, with thickness of 1-30 microm, were obtained as monolayers or laminates, containing chitosan-hydroxyapatite layers, separated by layers of pure chitosan. The hydroxyapatite nanoparticles showed preferred orientation in the chitosan matrix with the c-axis parallel to the substrate surface. The films showed corrosion protection of the Nitinol substrates in Ringer's physiological solutions. The feasibility of the fabrication of composite films containing hydroxyapatite and bioglass in the chitosan matrix has been demonstrated. The method offers the advantages of room temperature processing. The deposition mechanisms and possible applications of the films are discussed.

Polydopamine-polyethylene glycol-albumin antifouling coatings on multiple substrates.

Aqueous-based coatings using combinations of polydopamine (PDA) (as bioadhesive) and grafted polyethylene glycol (PEG) (as antifouling agent) have been reported to reduce biofouling on multiple material surfaces. However, the achievable PEG grafting density and antifouling performance are limited, leaving exposed PDA to provide sites for attachment of proteins and cells. In the present work, we investigate the polymerization of dopamine on three substrate materials, polycarbonate membrane (PC), polydimethyl siloxane (PDMS), and soda lime glass, to evaluate the utility of the PDA coatings for application to multiple materials. Additionally, we propose that the PDA-PEG method may be improved by "backfilling" with bovine serum albumin (BSA) as a blocker covering exposed PDA. AFM and ellipsometry studies revealed substantial differences in PDA thickness and roughness on each material despite their being modified under the same conditions. X-ray photoelectron spectroscopy (XPS) and water contact angle data revealed differences in PEG grafting on these materials as a consequence of varying PDA surface roughness, with the highest PEG coverage achieved on PC-PDA surfaces of intermediate roughness and lower PEG attachment on smoother PDMS-PDA surfaces. Fibrinogen adsorption experiments showed significantly less fouling on PDA-BSA surfaces compared to PDA-PEG for all three substrates, the larger BSA molecules presumably providing greater coverage of the PDA. On the PC and PDMS substrates, backfilling the PDA-PEG surfaces with BSA gave significant reductions in fibrinogen adsorption, with the lowest adsorption of 75 ng cm-2 achieved on PC-PDA-PEG/BSA.

Fibrinolytic properties of lysine-derivatized polyethylene in contact with flowing whole blood (Chandler loop model).

This article reports on the concept of a fibrinolytic surface based on the preferential adsorption of endogenous plasminogen from blood. Data are presented indicating that such a surface, when pretreated with tissue-type plasminogen activator (tPA), is able to dissolve nascent thrombus generated in contact with flowing whole blood. Polyethylene (PE) surfaces were modified by attaching a lysine-containing polymer using photochemical methods as reported previously (McClung et al., J Biomed Mater Res 2000;49:409-414). The lysine residues were bound chemically to the polymer via the alpha-amino groups leaving the epsilon-amino groups free (epsilon-Lys surface). Control surfaces were (a) unmodified PE, (b) PE modified with the coating polymer containing no lysine, and (c) PE modified with the polymer containing lysine bound via the epsilon-amino group. The materials in tubing form were evaluated in contact with nonanticoagulated flowing human whole blood in a modified Chandler Loop experiment. They were first treated with tPA to allow activation of adsorbed plasminogen to plasmin. It was found that thrombus formation was initiated within 15-25 min (depending on donor blood) on all surfaces, as indicated by the formation of platelet aggregates. On the controls (including the lysine-containing material in which the epsilon-amino group was used in the binding reaction) thrombogenesis continued till the tubing was occluded and blood flow ceased. On the epsilon-Lys surface, thrombogenesis was interrupted at various stages depending on the donor blood; in all cases any thrombus generated was dissolved within minutes. It was shown that thrombolysis was due to the fibrinolytic action of plasmin generated at the surface and not to plasmin formed by traces of tPA released into the blood. This work provides further evidence of the efficacy of this approach to the development of a fibrinolytic surface.

Fibrinogen adsorption and platelet lysis characterization of fluorinated surface-modified polyetherurethanes.

A polyetherurethane (PU) was modified using fluorinated surface-modifying macromolecules (SMMs). A double radiolabel method was used simultaneously to measure the number of adhered platelets ((51)Cr) and the quantity of adsorbed Fg ((125)I), in a cone-and-plate instrument. The objectives were to determine if adsorbed Fg levels correlated to platelet adhesion on the surfaces, and to assess if any reductions in platelet adhesion for the SMM-treated surfaces resulted from surface-induced platelet lysis, rather than changes directly related to lower platelet activation and attachment on the novel surfaces. Platelet lysis was determined from lactate dehydrogenase (LDH) and unbound (51)Cr released into plasma isolated from whole blood exposed to test materials. The corresponding Fg adsorption, evaluated under the same platelet adhesion conditions, did not account for the reduced platelet adhesion on the treated surfaces. LDH and (51)Cr platelet release were very low and indicated no statistically significant differences between the materials. It was therefore concluded that platelet lysis did not contribute to the reduction in platelet adhesion characteristic observed on the SMM-treated surfaces. More importantly, the work emphasizes that the platelet activation cannot be inferred to by assessing the quantity of fibrinogen as is commonly done in the literature. The finding suggests a much more complex mechanism of action for the SMM surface modifiers. On-going work is investigating other Fg parameters such as protein binding affinity and protein conformational state in order to establish the mechanism by which the fluorinated surface modifiers may be reducing platelet adhesion via intermediary changes in initial protein adsorption.

Protein adsorption to polyethylene glycol modified liposomes from fibrinogen solution and from plasma.

Unmodified and polyethylene glycol (PEG) modified neutral and negatively charged liposomes were prepared by freeze-thaw and extrusion followed by chromatographic purification. The effects of PEG molecular weight (PEG 550, 2000, 5000), PEG loading (0-15 mol%), and liposome surface charge on fibrinogen adsorption were quantified using radiolabeling techniques. All adsorption isotherms increased monotonically over the concentration range 0-3 mg/ml and adsorption levels were low. Negatively charged liposomes adsorbed significantly more fibrinogen than neutral liposomes. PEG modification had no effect on fibrinogen adsorption to neutral liposomes. An inverse relationship was found between PEG loading of negatively charged liposomes and fibrinogen adsorption. PEGs of all three molecular weights at a loading of 5 mol% reduced fibrinogen adsorption to negatively charged liposomes. Protein adsorption from diluted plasma (10% normal strength) to four different liposome types (neutral, PEG-neutral, negatively charged, and PEG-negatively charged) was investigated using gel electrophoresis and immunoblotting. The profiles of adsorbed proteins were similar on all four liposome types, but distinctly different from the profile of plasma itself, indicating a partitioning effect of the lipid surfaces. alpha2-macroglobulin and fibronectin were significantly enriched on the liposomes whereas albumin, transferrin, and fibrinogen were depleted compared to plasma. Apolipoprotein AI was a major component of the adsorbed protein layers. The blot of complement protein C3 adsorbed on the liposomes suggested that the complement system was activated.

Modification of liposomes with N-substituted polyacrylamides: identification of proteins adsorbed from plasma.

Liposomes prepared from DMPC (80%) and cholesterol (20%) were modified with a series of hydrophobically modified N-substituted polyacrylamides, namely, poly[N-isopropylacrylamide] (PNIPAM), poly[N,N-bis(2-methoxyethyl) acrylamide] (PMEAM), and poly[(3-methoxypropyl)acrylamide] (PMPAM). The hydrophobic group, N-[4-(1-pyrenylbutyl)-N-n-octadecylamine was attached to one end of the polymer chains to serve as an anchor for incorporation into the liposome bilayer. Liposome-polymer interactions were confirmed using fluorescence spectroscopy and chemical analysis. Microscopy revealed differences in aggregation tendency between unmodified and polymer-modified liposomes. Proteins adsorbed to liposome surfaces during exposure to human plasma were identified by immunoblot analysis. It was found that both unmodified and polymer-modified liposomes adsorb a wide variety of plasma proteins. Contact phase coagulation proteins, complement proteins, cell-adhesive proteins, serine protease inhibitors, plasminogen, antithrombin III, prothrombin, transferrin, alpha(2)-microglobulin, hemoglobin, haptoglobin and beta-lipoprotein as well as the major plasma proteins were all detected. Some differences were found between the unmodified and polymer-modified liposomes. The unmodified liposomes adsorbed plasminogen mainly as the intact protein, whereas on the modified liposomes plasminogen was present in degraded form. Also, the liposomes modified with PNIPAM in its extended conformation (below the lower critical solution temperature) appeared to adsorb less protein than those containing the 'collapsed' form of PNIPAM (above the LCST).

Fibrinogen surface distribution correlates to platelet adhesion pattern on fluorinated surface-modified polyetherurethane.

In previous work, it had been shown that platelet adhesion could be reduced by fluorinating surfaces with oligomeric fluoropolymers, referred to as surface-modifying macromolecules (SMMs). In the current study, two in vitro blood-contacting experiments were carried out on a polyetherurethane modified with three different SMMs in order to determine if altered platelet adhesion levels could be related to the pattern of adsorbed protein and more specifically to the manner in which fibrinogen (Fg) distribution occurs at the surface. In the first experiment, the materials were placed in whole human blood and the adherent platelets were viewed with high-resolution scanning electron microscopy (SEM). In a second experiment, the materials were incubated with human plasma with the absence of platelets. The plasma contained 5% fluorescent-Fg. The materials were then viewed with a fluorescence microscope and images were collected to define the distribution of high-density fluorescent-Fg areas. The SEM and fluorescent-Fg images were imported to Image Pro Plus imaging software to measure the area, length and circularity and a bivariate correlation test was conducted between the two sets of data. For area and length morphology parameters, there were high and significant correlations (r > 0.9, p < 0.05) between the platelets and Fg aggregates. The data suggest that the Fg distribution may serve as a predictor of platelet morphology/activation and provides insight into the non-thrombogenic character of biomaterials containing the fluorinated SMMs.

Patterns of adsorption of proteins from human plasma onto foreign surfaces.

The deposition of proteins on blood-contacting surfaces is known to be a determining factor in subsequent thromboembolic events. The composition of the protein layers and how they change with time are unknown. To generate information relevant to these questions, the quantities of albumin, fibrinogen and IgG adsorbed on seven surfaces from human plasma as a function of time were measured using a tracelabeling method. Materials studied include several segmented polyether-urethanes, glass, siliconized glass (SG), polystyrene (PS) and polyethylene (PE). Fibrinogen, surprisingly, was not adsorbed from plasma to any of the hydrophilic surfaces. On PE and SG adsorption passed through an early maximum (before 2 min) then declined to near zero. Only on PS was adsorption substantial and constant with time. Albumin was also not detected on the hydrophilic surfaces. IgG but was adsorbed substantially on the hydrophobic surfaces. IgG was detected on all surfaces, although in relatively low surface concentrations. These results suggest: 1. that the plasma itself interacts with initially adsorbed proteins, 2. that the role of fibrinogen adsorption in foreign-surface initiated thrombosis may need to be reevaluated and 3. that since the major plasma proteins are only minimally adsorbed, trace proteins may be important in blood-material interactions.

Effect of plasma dilution on adsorption of fibrinogen to solid surfaces.

The adsorption of various plasma proteins to three solid surfaces has been studied as a function of plasma concentration. Albumin adsorption on glass showed no dependence on plasma concentration and increased to a plateau value on both polyethylene and siliconized glass. Immunoglobulin G (IgG) adsorption showed no dependence on plasma concentration on any surface. Fibrinogen adsorption increased with plasma concentration and then decreased, the maximum occurring at about 1% normal plasma concentration and varying somewhat with the surface. On glass the kinetics of fibrinogen adsorption was dependent on plasma concentration: at concentrations less than the adsorption maximum, the kinetics was conventional, with adsorption increasing onto a plateau; at concentrations greater than the adsorption maximum, kinetics curves also showed maxima the position of which shifted to longer times as plasma concentration decreased. These data are interpreted in terms of competitive adsorption between fibrinogen and other, as yet unidentified species in plasma. The data reported are in general agreement with the model of Vroman (12) for plasma-surface interactions which holds that initially adsorbed fibrinogen is later replaced by high molecular weight kininogen (HMWK), the rate of replacement depending on the relative activity of the surface in promoting coagulation.

The effect of antithrombin III-independent thrombin inhibitors and heparin on fibrin accretion onto fibrin-coated polyethylene.

Prosthetic vascular grafts become coated with a layer of fibrin that contributes to graft thrombosis and occlusion. We compared the effect of antithrombin III-independent inhibitors of thrombin with heparin for their ability to prevent fibrin accretion onto a model of a vascular graft formed in vitro by coating polyethylene tubing with thrombin bound to a layer of polymerized fibrin. Equivalent antithrombin concentrations of heparin, D-Phe-Pro-Arg CH2Cl (PPACK), recombinant hirudin (r-hirudin), and Hirulog-1 were added to barium chloride-absorbed plasma containing radiolabelled fibrinogen. Whereas, PPACK and r-hirudin persistently inhibited fibrin accretion, the inhibition by heparin was transient. Hirulog-1 had no effect on early fibrin accretion and was actually associated with enhanced accretion at 30 min (control 11.7 +/- 2.0 micrograms fibrin/cm2; Hirulog-1, 18.4 +/- 3.5 micrograms fibrin/cm2, p < 0.001). Both Hirulog-1 and r-hirudin displaced radiolabelled thrombin from the fibrin surface. Whereas hirudin-thrombin complexes are stable, Hirulog-1 produces only transient inhibition of the displaced thrombin thereby accounting for the enhanced fibrin accretion with this anticoagulant. These studies show that the antithrombin III-independent inhibitors, r-hirudin and PPACK, are more effective inhibitors of fibrin accretion onto fibrin-coated polyethylene than heparin or Hirulog-1. In addition, they emphasize the importance of determining the ability of anticoagulants to displace thrombin from fibrin and to form stable thrombin-inhibitor complexes; lack of stability of thrombin-inhibitor complexes must be countered by levels of anticoagulant that are adequate to maintain its effectiveness.

Adsorption from plasma and buffer of single- and two-chain high molecular weight kininogen to glass and sulfonated polyurethane surfaces.

The adsorption of high molecular weight kininogen (HK) in its single-chain (SCHK) and two-chain (TCHK) forms from single protein solutions, plasma, and kininogen-deficient plasma, to glass and sulfonated polyurethane surfaces is reported. Using radiolabelling methods, it was found that in a single protein buffered system there was no difference in the adsorbed amounts of SCHK and TCHK over the concentration range 5-100 microg ml(-1) (similar to that in plasma). The adsorption of the two forms from normal plasma was also the same. However, immunoblots using an anti-HK antibody indicated that over the 2 h adsorption time, much of the SCHK present in the plasma was converted to TCHK: the band at 120 kD representative of intact SCHK disappeared, and bands at 56 and 46 kD representative of the heavy and light chains of TCHK were generated. To prevent conversion of SCHK to TCHK, the kallikrein inhibitor aprotinin (or in some cases a protease inhibitor cocktail), was added to the plasma in subsequent experiments. In addition, kininogen-deficient plasma was used (with either labelled SCHK or TCHK added) to avoid ambiguity in the tracer-population relationship. It was again found that there was no difference in the amounts of SCHK and TCHK adsorbed to glass and the sulfonated polyurethanes. The significance of these findings in relation to the reported anti-cell adhesion properties of adsorbed HK is discussed.

Adsorption of plasminogen from plasma to lysine-derivatized polyurethane surfaces.

The adsorption of plasminogen, the principal protein of the fibrinolytic pathway in blood, to a number of solid surfaces from plasma was investigated. This study forms part of a larger project to develop a fibrinolytic surface for blood-contacting applications. Polyurethanes incorporating lysine residues were developed in an attempt to promote selective adsorption of plasminogen from plasma through lysine-binding sites in the plasminogen molecule. The adsorption of plasminogen to these surfaces as well as to glass, 'conventional' polyurethanes and precursor sulphonated polyurethanes was investigated. Adsorption from citrated human plasma diluted with isotonic Tris buffer (pH 7.4) was measured under static conditions at room temperature using radioiodinated plasminogen. The following trends were observed. (1) Adsorption increases monotonically with increasing plasma concentration and there is no suggestion of transient adsorption (Vroman effect) on any of the surfaces studied. (2) Sulphonate groups appear to have a strong effect on plasminogen adsorption as was found previously for adsorption from buffer. (3) The lysine-derivatized material having the highest lysine content may show a slight increase in plasminogen binding affinity compared to its sulphonated precursor.

Adsorption of thrombin from buffer and modified plasma to polystyrene resins containing sulphonate and sulphamide arginyl methyl ester groups.

The adsorption of albumin and thrombin to insoluble modified polystyrene resins bearing sulphonate (PSSO3) and L-arginyl methyl ester groups (PAOM) was investigated in both purified and plasma systems. Radioiodinated proteins were used to follow adsorption in a 'minicolumn' experiment. Albumin adsorption was found to follow the Langmuir model and specific surface areas of the various resins were evaluated from plateau albumin adsorption data. The adsorption isotherms of thrombin both in buffer and in antithrombin III/fibrinogen-free plasma were also found to be Langmuir-like, and the quantities adsorbed at the isotherm plateaux are in the monolayer range. Analysis of the isotherms at 4 degrees and 37 degrees C for the purified system shows that adsorption is endothermic. Adsorption capacities in plasma remain high (30-50% of those in the purified system) despite competition from the other plasma proteins. These data confirm the strong affinity and selectivity of these resins for thrombin.

Protein and platelet interactions with thermally denatured fibrinogen and cross-linked fibrin coated surfaces.

In this work the hypothesis that a mature, cross-linked fibrin clot, pre-formed on a biomaterial, may be relatively nonthrombogenic was investigated. A cross-linked fibrin layer was formed on polyethylene which had been precoated with thermally denatured fibrinogen. Plasma protein adsorption and platelet interactions with the cross-linked fibrin and denatured fibrinogen surfaces were investigated. The adsorption of albumin, fibrinogen, and fibronectin from plasma was measured. For all three proteins, the cross-linked fibrin surface exhibited much higher levels of adsorption than either the thermally denatured fibrinogen or the polyethylene surface. Vroman peaks were observed for fibrinogen and fibronectin on polyethylene but not on the cross-linked fibrin and thermally denatured fibrinogen materials. In dilute plasma the thermally denatured fibrinogen surface showed considerable resistance to protein adsorption. However, at plasma concentrations greater than about 5% normal, this protein resistance was apparently lost. Platelet interactions (adhesion and release of granule constituents from adherent platelets) using suspensions of washed platelets in the presence of red cells were investigated at shear rates of 50, 300, and 525 s(-1) using a cone and plate apparatus. The levels of platelet adhesion on the different surfaces were in the order: adsorbed fibrinogen > cross-linked fibrin > thermally denatured fibrinogen = polyethylene. Platelets on the cross-linked fibrin surface also showed high levels of release indicating significant platelet activation. Scanning electron microscopic observations were in agreement with the platelet adhesion and release data, showing only a few (but well-spread) adherent platelets on the cross-linked fibrin surface.

Effect of C4-, C8- and C18-alkylation of poly(vinyl alcohol) hydrogels on the adsorption of albumin and fibrinogen from buffer and plasma: limited correlation with platelet interactions.

Polyvinyl alcohol (PVA) hydrogel was partially alkylated with short (C4), intermediate (C8) and long (C18) alkyl chains to test the hypothesis that an alkylated surface might promote enhanced interaction with albumin and thus exhibit low platelet thrombogenicity. PVA hydrogel was reacted with alkyl halides (C4, C8 or C18) and coated onto polyethylene. The effect of surface alkylation (extent of alkylation and alkyl chain length) on the adsorption of human serum albumin and fibrinogen to these surfaces was investigated in both buffer and plasma. Platelet interactions were investigated in vitro using flow cytometry methods. The maximum surface concentrations of albumin and fibrinogen adsorbed from buffer onto PVA and alkylated PVA were characteristic of monolayers. At low concentrations differences in adsorption among the surfaces appeared to be related to hydrophobicity as determined by dynamic advancing water contact angle, and to degree of alkylation as determined by angle dependent XPS analysis. Alkyl chain length dependence was not observed. Adsorption from plasma was considerably lower than from buffer, except for albumin on C8-PVA where monolayer adsorption was observed. Fibrinogen adsorption from plasma was similar on PVA, C8-PVA and C18-PVA, but was higher on C4-PVA. For albumin adsorption from plasma, the initial slope of the adsorption-concentration curve was highest for C18-PVA, suggesting higher albumin affinity despite the low degree of substitution of the C18-PVA material. These data suggest possible selectivity of the C18 alkylated PVA for albumin. Platelet studies showed that C4-PVA was the least platelet reactive (microparticle generation and P-selectin expression) of the alkyl derivatized materials.

Lysis of surface-localized fibrin clots by adsorbed plasminogen in the presence of tissue plasminogen activator.

The ability of plasminogen adsorbed from buffer onto sulphonated silica glass or lysine-derivatized silica glass to lyse fibrin I clots has been investigated. Clots were formed around the test surface by adding reptilase to fibrinogen solutions in which the surfaces were immersed. Tissue plasminogen activator (t-PA) was then added and the extent of clot lysis was determined by measuring the levels of the specific plasmin cleavage product of fibrinogen, B beta 1-42 peptide. The data indicate that in the presence of t-PA, B beta 1-42 generation per mole of bound plasminogen on the lysinized material is approximately two-fold higher than on the sulphonated material. It is concluded that a preformed clot may be lysed by adsorbed plasminogen in the presence of t-PA, and that clot lysis is significantly enhanced when the plasminogen is adsorbed via its lysine binding sites.

Lysine-derivatized polyurethane as a clot lysing surface: conversion of adsorbed plasminogen to plasmin and clot lysis in vitro.

Polyurethane surfaces to which lysine residues are immobilized by photochemical methods are proposed as a basis for clot lysing surfaces. The lysines are attached in such a way that the epsilon-amino and carboxyl groups are free. We showed previously that these surfaces, when placed in contact with plasma, adsorb only plasminogen and virtually no other proteins (McClung et al., J. Biomed. Mater. Res. 49 (2000) 409). In this communication, data based on a chromogenic substrate assay are presented showing that plasminogen adsorbed to these surfaces is readily converted to plasmin in the presence of tissue-plasminogen activator (t-PA). Moreover, the rate of activation on the surface is considerably greater than in solution. Experiments demonstrating the ability of these surfaces to dissolve fibrin clots are also reported. Surfaces exposed to plasma and then to t-PA were placed in citrated plasma. On recalcification, clotting was initiated, but the incipient clots were soon dissolved. On control surfaces (no lysine or lysine in which the epsilon-amino groups were not available) coagulation continued until a stable clot was formed. Similar observations were made when the plasma/t-PA exposed surfaces were placed in a pure fibrinogen solution and thrombin was added.

Immunoblot analysis of proteins associated with HEMA-MMA microcapsules: human serum proteins in vitro and rat proteins following implantation.

Human serum proteins and their fragments, associated with hydroxyethyl methacrylate-methyl methacrylate (HEMA-MMA) copolymer microcapsules, were characterized using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis. Capsules were incubated with serum for 1 week in vitro and then dissolved in ethanol to also precipitate the adsorbed protein. The precipitate was dissolved in 2% (w/v) SDS (the 'capsule eluate') to be assayed by electrophoresis. The majority of proteins probed for in the immunoblots were detected in the capsule eluates. These included fibronectin, plasminogen, IgG, vitronectin, Factor B, Factor H, Factor I, C3, but not beta-lipoprotein, fibrinogen, HMWK, or IgM. Complement activation fragments were detected in both the immunoblots of the capsule eluates and the medium containing serum without capsules. Thus, the adsorption of these fragments, formed independent of capsule presence, may be partially or completely responsible for the complement fragments associated with capsules. The prevention of complement activation by the addition of 5.8 mM EDTA, at the beginning of the week-long incubation, resulted in fewer low-molecular-weight C3 fragments associated with capsules. Rat proteins were also detected in immunoblots of the eluate of 'free-floating' capsules from the rat peritoneal cavity following implantation for 1 day using anti-human antibodies. Detected proteins included HMWK, fibrinogen, antithrombin III, transferrin, alpha1-antitrypsin, fibronectin, albumin, alpha2-macroglobulin, vitronectin, beta2-microglobulin, Factor B and Factor I. Rat fibrinogen, IgG, and complement C3 fragments were also detected in these immunoblots, but with monoclonal antibodies against the rat proteins.

Adsorption of immunoglobulin G and anti-factor VIII inhibitory antibody from haemophiliac plasma to derivatized polystyrenes.

Modified polystyrene resins containing sulphonate groups and tyrosyl sulphamide or tyrosyl methyl ester sulphamide groups have been investigated with respect to their potential for selective binding of anti-Factor VIII inhibitory antibodies from plasma. Adsorption of total immunoglobulin G and of a monoclonal antibody to Factor VIII was measured following addition of the radioiodinated proteins to normal plasma, plasma depleted of Factor VIII by adsorption on a resin coupled to anti-Factor VIII antibody, and haemophiliac plasma containing various levels of inhibitory anti-Factor VIII antibody. Depletion of anti-Factor VIII antibody from the haemophiliac plasmas by incubation with the resins was also measured by Bethesda assay. The modified resins and their corresponding unmodified "controls' showed similar binding of total immunoglobulin G. However, only resins containing either sulphonate or a combination of sulphonate and tyrosyl sulphamide groups showed evidence of selective adsorption of anti-Factor VIII antibody from plasma.