Adsorption of proteins from plasma to a series of hydrophilic-hydrophobic copolymers. I. Analysis with the in situ radioiodination technique

The blood compatibility challenge. Part 2: Protein adsorption phenomena governing blood reactivity

The adsorption of proteins is the initiating event in the processes occurring when blood contacts a "foreign" surface in a medical device, leading inevitably to thrombus formation. Knowledge of protein adsorption in this context has accumulated over many years but remains fragmentary and incomplete. Moreover, the significance and relevance of the information for blood compatibility are not entirely agreed upon in the biomaterials research community. In this review, protein adsorption from blood is discussed under the headings "agreed upon" and "not agreed upon or not known" with respect to: protein layer composition, effects on coagulation and complement activation, effects on platelet adhesion and activation, protein conformational change and denaturation, prevention of nonspecific protein adsorption, and controlling/tailoring the protein layer composition. STATEMENT OF SIGNIFICANCE: This paper is part 2 of a series of 4 reviews discussing the problem of biomaterial associated thrombogenicity. The objective was to highlight features of broad agreement and provide commentary on those aspects of the problem that were subject to dispute. We hope that future investigators will update these reviews as new scholarship resolves the uncertainties of today.

Fibrinogen adsorption to biomaterials

Fibrinogen (Fg) adsorption is an important mechanism underlying cell adhesion to biomaterials and was the major focus of the author's research career. This article summarizes our work on Fg adsorption, with citations of related work as appropriate. The molecular properties of Fg that promote adsorption and cell adhesion will be described. In addition, the adsorption behavior of Fg from buffer, binary solutions with other proteins, and blood plasma will be discussed, including the Vroman effect. Studies of platelet adhesion to surfaces preadsorbed with blood plasmas selectively deficient in Fg, vitronectin (Vn), fibronectin (Fn), or von Willebrand's factor (vWf) will be reviewed. These studies clearly showed a major role for Fg in platelet adhesion under static conditions and both Fg and vWf for adhesion from flowing suspensions, but no significant role for Vn or Fn. However, it was also shown that platelet adhesion was poorly correlated with the total amount of adsorbed Fg, but very well correlated with the binding of antibodies specific to the cell binding domains of Fg. A brief overview of nonfouling surfaces for prevention of Fg adsorption will be given. A more extensive discussion of structural changes in Fg after its adsorption is included, including changes detected with both physicochemical and biological methods. A short discussion of the state of the art of structural determination of adsorbed proteins with computational methods is also given. A final section identifies Fg adsorption as the single most important event determining the biocompatibility of implants in soft tissue and in blood. © 2018 Wiley Periodicals, Inc. J Biomed Mater Res Part A: 106A: 2777-2788, 2018.

Polyurethane-based microfluidic devices for blood contacting applications

Protein adsorption on PDMS surfaces poses a significant challenge in microfluidic devices that come into contact with biofluids such as blood. Polyurethane (PU) is often used for the construction of medical devices, but despite having several attractive properties for biointerfacing, it has not been widely used in microfluidic devices. In this work we developed two new fabrication processes for making thin, transparent and flexible PU-based microfluidic devices. Methods for the fabrication and bonding of microchannels, the integration of fluidic interconnections and surface modification with hydrophilic polyethylene oxide (PEO) to reduce protein adsorption are detailed. Using these processes, microchannels were produced having high transparency (96% that of glass in visible light), high bond strength (326.4 kPa) and low protein adsorption (80% reduction in fibrinogen adsorption vs. unmodified PDMS), which is critical for prevention of fouling. Our findings indicate that PEO modified PU could serve as an effective alternative to PDMS in blood contacting microfluidic applications.

Impact of fluorescent probes on albumin sorption profiles to ophthalmic biomaterials

Bovine serum albumin (BSA) was conjugated to three organic fluorescent probes, 5-(4,6-dichloro-s-triazin-2-ylamino)fluorescein hydrochloride (DTAF), Rhodamine B isothiocyanate (RITC), and Lucifer yellow VS (LY). The protein sorption profile to one pHEMA-based (etafilcon A) and three silicone hydrogel (SH) contact lens types (lotrafilcon B, balafilcon A and senofilcon A) was determined using confocal laser scanning microscopy. In addition, all lenses were incubated in dye solutions containing the fluorescent probe alone; and in a separate experiment BSA accumulation was quantified using radiolabeling. The different fluorescent conjugates showed similar sorption profiles for the pHEMA-based lens, but marked differences for all SH lenses. Lotrafilcon B accumulated more protein on the surface as compared to the matrix, independent of the fluorescent probe used for conjugation. Protein sorption varied for senofilcon A, with DTAF-BSA sorbing primarily to the surface region, while the other conjugates penetrated in equal amounts into the matrix. Balafilcon A exhibited smaller differences between conjugates, with LY-BSA allowing the protein to fully penetrate the matrix, while the other conjugates showed minor surface adsorption. Sorption curves of unbound dyes were often similar compared to the conjugated results. BSA profiles to pHEMA-based and silicone hydrogel lenses were highly dependent on the fluorescent probe used and none of the probes accurately reflected quantitative protein levels for the lens materials investigated.

Kinetics ofIn VitroLysozyme Deposition on Silicone Hydrogel, PMMA, and FDA Groups I, II, and IV Contact Lens Materials

We sought to compare the kinetics of in vitro lysozyme deposition on silicone hydrogel (SH), polymethyl methacrylate (PMMA), and FDA groups I, II, and IV contact lenses. Lenses were incubated in 125I-labeled lysozyme for time periods ranging from 1 hr to 28 days, and radioactive counts were determined. SH lenses and PMMA deposited less lysozyme than conventional hydrogel lenses (p < 0.05). Lysozyme accumulation on group IV lenses reached a maximum on the seventh day and then plateaued, whereas on groups I, II, and SH lenses, deposition continued to increase across all time periods, reiterating that kinetics of lysozyme deposition is highly material dependent.

Tetraglyme coatings reduce fibrinogen and von Willebrand factor adsorption and platelet adhesion under both static and flow conditions

Previous studies have showed that radio-frequency plasma deposited tetraglyme coatings greatly reduced fibrinogen adsorption (Gamma(Fg)) from highly diluted plasmas (0.1 and 1%) and subsequent platelet adhesion under static conditions. In this study, the protein resistant properties of tetraglyme were re-examined with high-concentration plasma, and subsequent platelet adhesion was measured under both static and flow conditions. The resistance of tetraglyme to vWf adsorption (Gamma(vWf)) and the role of vWf in platelet adhesion under flow were also investigated. Gamma(Fg) and Gamma(vWf) were measured with (125)I radiolabeled proteins. Flow studies were done at shear rates of 50 or 500 s(-1) by passing a platelet/red cell suspension through a GlycoTech flow chamber. When adsorbed from a series of increasing plasma concentrations, the adsorption of both proteins to tetraglyme increased steadily, and did not show a peak at intermediate dilutions, i.e., there was no Vroman effect. When plasma concentration was less than 10%, the tetraglyme surface was highly nonfouling, exhibiting ultralow Gamma(Fg) (less than 5 ng/cm(2)) and extremely low platelet adhesion under both static and flow conditions. However, when the adsorption was done from 100% plasma, Gamma(Fg) was much higher ( approximately 85 ng/cm(2)), indicating that tetraglyme surface may not be sufficiently protein-resistant in the physiological environment. To correlate platelet adhesion under flow with Gamma(Fg) and Gamma(vWf), a series of tetraglyme surfaces varying in ether content and protein adsorption was created by varying deposition power. On these surfaces, platelet adhesion at low shear rate depended only on the amount of Gamma(Fg), but under high shear, both Gamma(Fg) and Gamma(vWf) affected platelet adhesion. In particular, it was found that Gamma(vWf) must be reduced to less than 0.4 ng/cm(2) to achieve ultra low platelet adhesion under high shear.

Fibrinogen and von Willebrand factor mediated platelet adhesion to polystyrene under flow conditions

The roles of adsorbed fibrinogen (Fg) and von Willebrand factor (VWF) in mediating platelet adhesion to synthetic surfaces under flow were investigated using polystyrene (PS) as a model hydrophobic surface. We measured platelet adhesion to PS pre-adsorbed with Fg, VWF, normal plasma, afibrinogenemic plasma, VWF-deficient plasma and deficient plasmas with various concentrations of added Fg or VWF. Platelets in a red blood cell suspension were passed through a flow chamber at either low (50 or 100 s(-1)) or high (500 or 1000 s(-1)) shear. Adhesion to PS pre-adsorbed with afibrinogenemic plasma was very low under both low and high shear conditions, but was restored in a dose-dependent manner with addition of Fg. Less than 20 ng/cm(2)of adsorbed Fg was sufficient to support full-scale platelet adhesion under flow. At high shear rate, platelet adhesion on PS pre-adsorbed with VWF-deficient plasma was much less than on PS pre-adsorbed with normal plasma, but adhesion to PS pre-adsorbed with VWF-deficient plasma with added VWF was very similar to adhesion to PS pre-adsorbed with normal plasma. At low shear, adhesion to PS pre-adsorbed with VWF-deficient plasma was the same as on PS pre-adsorbed with normal plasma. As little as 1 ng/cm(2) of VWF adsorbed from plasma made platelet adhesion higher under high shear than under low shear. The effects of adsorbed Fg and VWF on the morphologies of platelets that adhered from suspensions flowing at high shear rates were also investigated. The lack of either Fg or VWF resulted in marked decreases in the extent of platelet spreading. Real-time observation of platelet adhesion under an epifluorescent microscope showed that platelets adhered to the surface in a linear pattern aligned in the direction of flow under high shear conditions.

Extraction Efficiency of an Extraction Buffer Used to Quantify Lysozyme Deposition on Conventional and Silicone Hydrogel Contact Lens Materials

PURPOSE

Extracting lysozyme from Food and Drug Administration group IV etafilcon lenses by using 0.2% trifluoroacetic acid and acetonitrile (TFA/ACN) is a well-established procedure. TFA/ACN has been the extraction buffer of choice for extracting proteins from silicone hydrogel contact lenses. The purpose of this study was to determine the efficiency of TFA/ACN in extracting lysozyme from silicone hydrogel and etafilcon lenses by using an in vitro model.

METHODS

ACUVUE 2, Focus NIGHT & DAY, O2 Optix, PureVision, and ACUVUE Advance lenses were incubated in simple lysozyme solution and a complex artificial tear solution consisting of multiple tear components containing lysozyme labeled with iodine 125. All the silicone hydrogel lenses were incubated for 28 days, whereas the ACUVUE 2 lenses were incubated for 7 days at 37 degrees C with constant rotation. After the incubation period, radioactive counts were determined, and the lenses were placed in an appropriate volume of the buffer for 24 hours in darkness. The lenses were removed from the buffer, and radioactive counts were determined again.

RESULTS

Extraction efficiencies for lysozyme from the artificial tear solution were 97.2% +/- 1.2% for ACUVUE 2, 64.3% +/- 6.2% for Focus NIGHT & DAY, 62.5% +/- 5.6% for O2 Optix, 53.5% +/- 5.8% for PureVision, and 89.2% +/- 3.4% for ACUVUE Advance. Results were similar for the lysozyme extracted after incubating in the simple lysozyme solution.

CONCLUSIONS

TFA/ACN is extremely efficient at extracting lysozyme deposited on etafilcon lenses. However, it does not extract all the lysozyme deposited on silicone hydrogel lenses, and alternative extraction procedures should be sought.

Radiotracer measurements of protein mass transfer: Kinetics in ion exchange media

We describe a method to measure protein mass transfer kinetics in ion exchange adsorbents for preparative chromatography based on the use of radioactively labeled protein. The method was developed and evaluated using lysozyme as a test protein with the three commercial strong-acid cation exchangers SP-Sepharose-FF, SP-Sepharose-XL, and S-HyperD. Iodination with 125I was used to label the protein, which was added in trace amounts (approximately 0.1%) to an unlabeled protein solution. The solution was recirculated through a shallow bed of the adsorbent particles and the radioactivity accumulated in the bed measured with a gamma-counter as a function of time. Radiotracer-based kinetics measurements were found to be in good agreement with results obtained with a conventional shallow-bed technique, provided that freshly labeled protein solutions were used. The method has advantages in terms of simplicity, ability to deal with adsorption from complex mixtures, and the potential for measurements under tracer diffusion conditions. Kinetics results obtained for the three different stationary phases were generally consistent with previous studies. Protein mass transfer can be described by a pore diffusion model with a nearly salt-independent pore diffusivity for SP-Sepharose-FF and by a homogeneous diffusion model with a saltindependent adsorbed phase diffusivity for S-HyperD. However, it appears that a more complex model, accounting for parallel pore and surface diffusion, is needed to describe protein mass transfer in SP-Sepharose-XL. The modeling results were found to be correlated with the apparent pore sizes determined by inverse SEC.

Hemocompatibility of low-friction boron–carbon–nitrogen containing coatings

Mechanical heart valves are exposed to extreme mechanical demands, which require a surface showing not only nonhaemostatic properties, but also wear resistance and low friction. As alternative to different forms of amorphous carbon (a-C), so-called diamond-like carbon (DLC), the suitability of boron carbonitride (BCN) coatings is tested here for hemocompatible coatings. They have similar mechanical properties like a-C surfaces, but superior chemical stability at ferrous substrates or counterparts. BCN films with different nitrogen content were compared with hydrogenated a-C films regarding their mechanical properties, surface energy, adsorption of albumin and fibrinogen, blood platelet adherence, and activation of the contact system of the clotting cascade and kinin system. Similar mechanical properties and biological response have been found in the BCN films with respect to a-C, indicating the potential of these coatings for biomedical applications. The increase in the crystallinity and tribological properties of the BCN samples with a higher incorporation of N was also followed by a lower protein adsorption and low activation of the contact system, but an increased adherence of thrombocytes.

Proteomic analysis of protein adsorption: Serum amyloid P adsorbs to materials and promotes leukocyte adhesion

Serum and plasma protein adsorption on materials was analyzed using gel electrophoresis and ion trap mass spectrometry. Following incubation of polypropylene, polyethylene terephthalate (PET), or polydimethylsiloxane (PDMS) with 5% serum for longer than 4 h, we found unexpectedly high amounts of the pentraxin serum amyloid P. It was previously shown that serum amyloid P is constitutively expressed in humans, functions as an opsonin, and interacts with the Fcgamma receptors on leukocytes. We demonstrate that serum amyloid P adsorbed to tissue culture polystyrene, PDMS, and PET promotes the adhesion of granulocytes and monocytes in the presence of calcium. The methods developed for these studies may be useful for the large-scale study of protein adsorption and do not rely on radiolabeling or the availability of antibodies.

The role of adsorbed fibrinogen in platelet adhesion to polyurethane surfaces: A comparison of surface hydrophobicity, protein adsorption, monoclonal antibody binding, and platelet adhesion

Ten specially synthesized polyurethanes (PUs) were used to investigate the effects of surface properties on platelet adhesion. Surface composition and hydrophilicity, fibrinogen (Fg) and von Willebrand's factor (vWf) adsorption, monoclonal anti-Fg binding, and platelet adhesion were measured. PUs preadsorbed with afibrinogenemic plasma or serum exhibited very low platelet adhesion, while adhesion after preadsorption with vWf deficient plasma was not reduced, showing that Fg is the key plasma protein mediating platelet adhesion under static conditions. Platelet adhesion to the ten PUs after plasma preadsorption varied greatly, but was only partially consistent with Fg adsorption. Thus, while very hydrophilic PU copolymers containing PEG that had ultralow Fg adsorption also had very low platelet adhesion, some of the more hydrophobic PUs had relatively high Fg adsorption but still exhibited lower platelet adhesion. To examine why some PUs with high Fg adsorption had lower platelet adhesion, three monoclonal antibodies (mAbs) that bind to sites in Fg thought to mediate platelet adhesion were used. The antibodies were: M1, specific to gamma-chain C-terminal; and R1 and R2, specific to RGD containing regions in the alpha-chain N- and C-terminal, respectively. Platelet adhesion was well correlated with M1 binding, but not with R1 or R2 binding. When these mAbs were incubated with plasma preadsorbed surfaces, they blocked adhesion to variable degrees. The ability of the R1 and R2 mAbs to partially block adhesion to adsorbed Fg suggests that RGD sites in the alpha chain may also be involved in mediating platelet adhesion and act synergistically with the C-terminal of the gamma-chain.

Characterization and analysis of osteopontin-immobilized poly(2-hydroxyethyl methacrylate) surfaces

A wide array of technologies exist for the characterization and quantification of molecules present at surfaces. We have used several of these experimental and instrumental techniques for the analysis of a novel biomaterial surface. Osteopontin, an extracellular matrix molecule involved in wound-healing processes, has been chosen as a relevant model protein to immobilize onto poly(2-hydroxyethyl methacrylate) [poly(HEMA)]. Electron spectroscopy for chemical analysis and time-of-flight secondary ion mass spectrometry were used to verify the surface chemistry and the presence of protein. Iodination of osteopontin yielded quantitative data supportive of dose-dependent immobilization. Enzyme-linked immunosorbent assay was also used to investigate the presence of osteopontin on poly(HEMA). Finally, the cell adhesive properties of immobilized osteopontin were confirmed by using a bovine aortic endothelial cell adhesion assay. The use of multiple tools to characterize the many facets of a biomaterial surface will undoubtedly improve our understanding of the surface and facilitate the amelioration of in vivo performance.

The influence of protein adsorption and surface modifying macromolecules on the hydrolytic degradation of a poly(ether-urethane) by cholesterol esterase

Previous investigations have demonstrated that the inflammatory cell derived enzyme, cholesterol esterase (CE) could degrade polyurethanes (PUs) by hydrolyzing ester and urethane bonds. Studies that have investigated the development of protective coatings for PUs have reported that the polymer degradation of polyester-urethanes (PESUs) can be reduced with the use of fluorine containing surface modifying macromolecules (SMMs). Since these latter studies were carried out in the presence of relatively pure enzyme, it has not been shown if SMMs would still provide an enhanced inhibitory effect if surfaces were pre-exposed to plasma proteins. This would be more representative of the in vivo scenario since protein adsorption would occur before the appearance of monocyte-derived macrophages which would be a primary source of esterase activities. The current investigation has focused on studying the influence of fibrinogen (Fg) as a simple model of protein adsorption in order to assess the effect of CE in combination with protein on polyether-urethane (PEU) surfaces. The materials were prepared with and without SMMs, and were pre-coated with Fg prior to carrying out biodegradation studies. The pre-adsorption of Fg onto the modified and non-modified surfaces provided a significant delay in the hydrolytic action of CE onto the PEU substrates. However, the effect was gone by 70 days and by the 126th day of incubation, both Fg coated and non-Fg coated groups had the same level of degradation. The difference between Fg coated and non-coated substrates was much smaller for materials containing SMMs. In addition, the pre-adsorption of Fg did not alter the SMMs' ability to provide a more biostable surface over the 4 month incubation period.

Platelet adhesion to polystyrene-based surfaces preadsorbed with plasmas selectively depleted in fibrinogen, fibronectin, vitronectin, or von Willebrand's factor

Four plasma proteins have been shown to be able to mediate platelet adhesion to synthetic materials when they are adsorbed as purified proteins: fibrinogen (Fg), fibronectin (Fn), vitronectin (Vn), and von Willebrand factor (vWF). Among them, Fg is thought to play a leading role in mediating platelet adhesion to plasma-preadsorbed biomaterials, but this has been established for only three types of materials so far in our laboratory. Furthermore, the role of Fn, Vn, and vWF in mediating platelet adhesion to plasma-preadsorbed surfaces is still unclear. The aim of the current study was to assess the importance of Fg, Fn, Vn, and vWF in mediating platelet adhesion to a series of polystyrene-based surfaces. The strategy applied in the present investigation was to compare platelet adhesion to surfaces preadsorbed with normal plasma, plasma selectively depleted in Fn or Vn or both Fn and Vn, plasma from donors who were genetically deficient in vWF, and serum. Few platelets adhered to the surfaces preadsorbed with serum, whereas depletion of Fn, Vn, or vWF from plasma did not decrease platelet adhesion significantly. Replenishment of exogenous Fg to serum before protein adsorption restored platelet adhesion to the surfaces, suggesting that Fg was the major plasma protein that mediated platelet adhesion. Also, we found that a surface density of adsorbed Fg far below the amount that usually adsorbs to synthetic surfaces was sufficient to support full-scale platelet adhesion.

The biocompatibility of novel starch-based polymers and composites: in vitro studies

Studies with biodegradable starch-based polymers have recently demonstrated that these materials have a range of properties. which make them suitable for use in several biomedical applications, ranging from bone plates and screws to drug delivery carriers and tissue engineering scaffolds. The aim of this study was to screen the cytotoxicity and evaluate starch-based polymers and composites as potential biomaterials. The biocompatibility of two different blends of corn-starch, starch ethylene vinyl alcohol (SEVA-C) and starch cellulose acetate (SCA) and their respective composites with hydroxyapatite (HA) was assessed by cytotoxicity and cell adhesion tests. The MTT assay was performed with the extracts of the materials in order to evaluate the short-term effect of the degradation products. The cell morphology of L929 mouse fibroblast cell line was also analysed after direct contact with polymers and composites for different time periods and the number of cells adhered to the surface of the polymers was determined by quantification of the cytosolic lactate dehydrogenase (LDH) activity. Both types of starch-based polymers exhibit a cytocompatibility that might allow for their use as biomaterials. SEVA-C blends were found to be the less cytotoxic for the tested cell line, although cells adhere better to SCA surface. The cytotoxicity test also revealed that SCA and SEVA-C composites have a similar response to the one obtained for SCA polymer. Scanning electron microscopy (SEM) analysis showed that cells were much more spread on the SCA polymer and LDH measurements showed a higher number of cells on this surface.

Cellulose phosphates as biomaterials. In vivo biocompatibility studies

Femoral implantation of regenerated cellulose hydrogels revealed their biocompatibility, but a complete osseointegration could not be observed. Phosphorylation was therefore envisaged as the means to enhance cellulose bioactivity. In vitro studies showed that regenerated cellulose hydrogels promote bone cells attachment and proliferation but do not mineralize in acellular simulated physiological conditions. On the contrary, phosphorylated cellulose has shown an opposite behavior, by inducing the formation of a calcium phosphate layer in simulated physiological conditions, but behaving as a poor substrate for bone cells attachment and proliferation. In order to investigate the in vivo behavior of these materials, and assess the influence of mineralization induction ability vs. bone cells compatibility, unmodified and phosphorylated cellulose hydrogels were implanted in rabbits for a maximum period of 6 months and bone regeneration was investigated. Despite the difficulties arising from the retraction of cellulose hydrogels upon dehydration during the preparation of retrieved implants, histological observations showed no inflammatory response after implantation, with bone intra-spongious regeneration of cells and the integration of the unmodified as well as the phosphorylated cellulose implants. After a maximum implantation period of 6 months, histological observations, histomorphometry and the measurement of the amount of 45Ca incorporated in the surrounding tissue indicated a slightly better osseointegration of phosphorylated cellulose, although no significant differences between the two materials were found.

Plasma Protein Adsorption and Platelet Adhesion on Poly

In this work we describe experiments designed to understand the blood compatibility and resistance to platelet adhesion of poly[bis(trifluoroethoxy)phosphazene] (Acta Polymerica 36, 627 (1985)) (PTFEP) coated surfaces. We compare quantitative in vitro protein adsorption measurements using enzyme linked immunosorbent assays (ELISA) and platelet adherence tests on PTFEP with other organic surfaces and hydroxylated glass. Compared to some materials of medical interest (polymethylmethacrylate and silicone) and other materials (hydroxylated glass, aldehyde-, alkyl-, or amino-terminated surfaces) exhibiting a wide range of physical properties, PTFEP showed the highest human serum albumin adsorption and the lowest adsorption of fibrinogen and fibronectin. These proteins are related to thrombus formation and cellular attachment, respectively. Coagulation-stimulating proteins are predominantly bound reversibly on PTFEP and do not appear to denaturate to the extent found on the other surfaces. Copyright 1998 Academic Press. Copyright 1998Academic Press

Culture materials affect ex vivo expansion of hematopoietic progenitor cells

Ex vivo expansion of hematopoietic cells is important for applications such as cancer treatment, gene therapy, and transfusion medicine. While cell culture systems are widely used to evaluate the biocompatibility of materials for implantation, the ability of materials to support proliferation of primary human cells in cultures for reinfusion into patients has not been addressed. We screened a variety of commercially available polymer (15 types), metal (four types), and glass substrates for their ability to support expansion of hematopoietic cells when cultured under conditions that would be encountered in a clinical setting. Cultures of peripheral blood (PB) CD34+ cells and mononuclear cells (MNC) were evaluated for expansion of total cells and colony-forming unit-granulocyte monocyte (CFU-GM; progenitors committed to the granulocyte and/or monocyte lineage). Human hematopoietic cultures in serum-free medium were found to be extremely sensitive to the substrate material. The only materials tested that supported expansion at or near the levels of polystyrene were tissue culture polystyrene, Teflon perfluoroalkoxy, Teflon fluorinated ethylene propylene, cellulose acetate, titanium, new polycarbonate, and new polymethylpentene. MNC were less sensitive to the substrate materials than the primitive CD34+ progenitors, although similar trends were seen for expansion of the two cell populations on the substrates tested. CFU-GM expansion was more sensitive to substrate materials than was total cell expansion. The detrimental effects of a number of the materials on hematopoietic cultures appear to be caused by protein adsorption and/or leaching of toxins. Factors such as cleaning, sterilization, and reuse significantly affected the performance of some materials as culture substrates. We also used PB CD34+ cell cultures to examine the biocompatibility of gas-permeable cell culture and blood storage bags and several types of tubing commonly used with biomedical equipment. While many of the culture bag materials gave satisfactory results, all of the tubing materials severely inhibited total cell and CFU-GM expansion. Taken together, our results show that many materials approved for blood contact or considered biocompatible are not suitable for use with hematopoietic cells cultured in serum-free medium. As hematopoietic cultures are scaled up for a variety of clinical applications, it will be essential to carefully examine the biocompatibility of all materials involved.

Influence of block copolymers on the adsorption of plasma proteins to microspheres

SDS-PAGE in combination with densitometry has been used to evaluate the adsorption of plasma and serum proteins to polystyrene microspheres (PS) coated with block copolymers of the poloxamer and poloxamine series. The protein-resistant nature of coated PS was demonstrated for these systems when incubated in dilutions of plasma and serum. The total amount of protein and the type of proteins adsorbed were dependent on the plasma and serum incubation concentration used. At 0.3% serum concentrations the total amount of protein adsorbed was found to be related to the polyoxyethylene (PEO) chain length of the block copolymer, whilst at 0.3% or 50% plasma concentrations a relationship was shown between the polyoxypropylene (PPO) chain and the plasma protein adsorption for the range of block copolymers studied. Immunoblotting studies revealed the adsorption of immunoglobulin G, complement C3, transferrin and fibronectin to all microspheres previously incubated in 50% serum and plasma, whilst fibrinogen was also adsorbed after incubation in 50% plasma; with similar quantities of each protein adsorbed to PS and block copolymer-coated PS.

Postadsorptive transitions in fibrinogen adsorbed to polyurethanes: changes in antibody binding and sodium dodecyl sulfate elutability

Residence time-dependent changes in fibrinogen after adsorption to six different polyurethanes were examined by measuring polyclonal antifibrinogen binding to the adsorbed protein. The amount of adsorbed fibrinogen that could be eluted by sodium dodecyl sulfate (SDS) was also measured. Baboon fibrinogen was first adsorbed from dilute plasma to the polymers, which were then stored in either buffer or buffered albumin solution prior to testing. Subsequently, the amount of antifibrinogen bound by the adsorbed fibrinogen was measured using a direct enzyme linked immunosorbent assay (ELISA). Alternatively, the surface with the adsorbed fibrinogen was soaked in a 3% SDS solution, and the amount of retained 125I-radiolabeled fibrinogen was measured. With increasing residence time, decreases in both antibody binding and the SDS elutability of the adsorbed fibrinogen occurred, but the rate of change was dependent on the polyurethane to which the fibrinogen was adsorbed. In addition, the antibody binding per unit of adsorbed fibrinogen, when measured immediately after the adsorption step, varied by approximately a factor of 3 among the various polyurethanes. When the protein-coated surfaces were stored in buffered albumin solution rather than buffer, the decrease in the reactivity of fibrinogen with residence time did not occur on some of the surfaces. This study shows that the chemical properties of the adsorbing surface influence the rate at which adsorbed fibrinogen undergoes change. The significance of the polymer-dependent changes in adsorbed fibrinogen with respect to blood reactions with polymers is discussed.

On the relative importance of specific and non-specific approaches to oral microbial adhesion

In this paper, it is suggested that specificity and non-specificity in (oral) microbial adhesion are different expressions for the same phenomena. It is argued that the same basic, physicochemical forces are responsible for so-called 'non-specific' and 'specific' binding and that from a physico-chemical point of view the distinction between the two is an artificial one. Non-specific interactions arise from Van der Waals and electrostatic forces and hydrogen bonding, and originate from the entire cell. A specific bond consists of a combination of the same type of Van der Waals and electrostatic forces and hydrogen bonding, now originating from highly localized chemical groups, which together form a stereochemical combination. The absence or presence of specific receptor sites on microbial cell surfaces must therefore be reflected in the overall, non-specific surface properties of cells as well. This point is illustrated by showing that glucan-binding lectins on mutans streptococcal strains may determine the pH dependence of the zeta potentials of these cells. When studying microbial adhesion, a non-specific approach may be better suited to explain adhesion to inert substrata, whereas a specific approach may be preferred in case of adhesion to adsorbed protein films. Adhesion is, however, not as important in plaque formation in the human oral cavity as is retention, because low shear force periods, during which adhesion presumably occurs, are followed by high shear force periods, during which adhering cells must withstand these detachment forces. Evidence is provided that such detachment will be through cohesive failure in the pellicle mass, the properties of which are conditioned by the overall, non-specific substratum properties. Therefore, in vivo plaque formation may be more readily explained by a non-specific approach.

Influence of substratum wettability on the strength of adhesion of human fibroblasts

To determine the strength of adhesion and the detachment mechanisms of fibroblasts from substrata with different wettability, the behaviour of adhered cells was studied in a parallel-plate flow chamber during exposure to shear. Adhered cells were observed in situ, i.e. in the flow chamber, by phase-contrast microscope and images were analysed semiautomatically. Detachment was found to be dependent both on shear stress and time, although a critical shear stress can be found below which no detachment occurs. On all substrata, cells round up before detachment and are approximately spherical immediately before detachment. The strength of adhesion calculated ranged from 0.6-3.5 x 10(-10) N per cell on FEP-Teflon (the least wettable material included) to 9.4 x 10(-9) N per cell for glass (the most wettable). Ease of detachment seemed to decrease with increasing wettability. However, cells reacted more strongly with tissue culture polystyrene (TCPS) than expected on the basis of its wettability, probably due to surface chemistry.

Human serum albumin as a probe for surface conditioning (opsonization) of block copolymer-coated microspheres

The adsorption of human serum albumin to polystyrene microspheres sterically stabilized with block copolymers, was investigated using photon correlation spectroscopy and laser doppler anenometry. The block copolymers used were non-ionic surfactants of the poloxamer and poloxamine series made of polyoxyethylene and polyoxypropylene chains. Photon correlation spectroscopy and laser doppler anenometry showed that the coating reduced the adsorption of the protein to the polystyrene microspheres surface. Quantitative studies using 125I-labelled human serum albumin and sodium dodecyl sulphate-polyacrylamide gels (in combination with densitometry), were employed to evaluate the adsorption of human serum albumin to uncoated and coated polystyrene microspheres. They confirmed that the hydrophilic polyoxyethylene steric layer, created by coating with the block copolymers, reduced the adsorption of human serum albumin. Moreover, the amount of human serum albumin adsorbed was related to the polyoxyethylene content of the block copolymers.

Protein adsorption to poly(ethylene oxide) surfaces

Surfaces containing poly(ethylene oxide) (PEO) are interesting biomaterials because they exhibit low degrees of protein adsorption and cell adhesion. In this study different molecular weight PEO molecules were covalently attached to poly(ethylene terephthalate) (PET) films using cyanuric chloride chemistry. Prior to the PEO immobilization, amino groups were introduced onto the PET films by exposing them to an allylamine plasma glow discharge. The amino groups on the PET film were next activated with cyanuric chloride and then reacted with bis-amino PEO. The samples were characterized by scanning electron microscopy, water contact angle measurements, gravimetric analysis, and electron spectroscopy for chemical analysis (ESCA). The adsorption of 125I-labeled baboon fibrinogen and bovine serum albumin was studied from buffer solutions. Gravimetric analysis indicated that the films grafted with the low-molecular-weight PEO contained many more PEO molecules than the surfaces grafted with higher-molecular-weight PEO. The high-molecular-weight PEO surfaces, however, exhibited greater wettability (lower water contact angles) and less protein adsorption than the low-molecular-weight PEO surfaces. Adsorption of albumin and fibrinogen to the PEO surfaces decreased with increasing PEO molecular weight up to 3500. A further increase in molecular weight resulted in only slight decreases in protein adsorption. Protein adsorption studies as a function of buffer ionic strength suggest that there may be an ionic interaction between the protein and the allylamine surface. The trends in protein adsorption together with the water contact angle results and the gravimetric analysis suggest that a kind of "cooperative" water structuring around the larger PEO molecules may create an "excluded volume" of the hydrated polymer coils. This may be an important factor contributing to the observed low protein adsorption behavior.

Postadsorptive transitions in fibrinogen adsorbed to biomer: changes in baboon platelet adhesion, antibody binding, and sodium dodecyl sulfate elutability

Residence-time-dependent changes in fibrinogen after its adsorption to Biomer were examined by measuring platelet adhesion and antibody binding to the adsorbed protein, and the amount of adsorbed fibrinogen which could be eluted by sodium dodecyl sulfate (SDS). Baboon fibrinogen was first adsorbed (from either pure solution or dilute plasma) to Biomer, which was then stored in either buffer or buffered albumin solution prior to testing. Subsequently, the adherent protein layer was either probed for fibrinogen capable of mediating platelet adhesion using 111In radiolabeled, washed platelet suspensions under both static and shearing conditions, or for fibrinogen capable of binding antibody using a direct enzyme linked immunosorbent assay (ELISA). Alternatively, the surface with the adsorbed protein layer was soaked in a 3% SDS solution, and the amount of 125I radiolabeled fibrinogen retained was measured. Decreases in platelet and antibody binding, and in the SDS elutability of the adsorbed fibrinogen after it was stored in buffer were detected, although different rates of decrease were observed for each method. When the protein-coated surfaces were stored in buffered albumin solution rather than buffer, the decrease in the reactivity of fibrinogen was prevented. While each of the three assays measures a different property of adsorbed fibrinogen, this study suggests that the adherent protein undergoes time dependent conformational changes which render it less reactive toward platelets and antibodies, and more resistant to elution by SDS.

Secretory IgA adsorption and oral streptococcal adhesion to human enamel and artificial solid substrata with various surface free energies

In this paper, secretory IgA adsorption from a single component sIgA solution and from human whole saliva onto human enamel and artificial solid substrata with various surface free energies was studied as a function of time. ELISA indicated that screening or displacement of adsorbed sIgA by other salivary proteins occurred only on low surface free energy substrata, not on high surface free energy substrata such as enamel. In addition, the adhesion of three oral streptococcal strains (Streptococcus mitis BMS, S. sanguis 12, and S. mutans NS), also having widely different surface free energies, to sIgA-coated surfaces was studied. The adhesion of all three streptococcal strains was significantly reduced in the presence of a sIgA coating. However, ranking the adhesion data with respect to the various substrata revealed a similar order to that in the case of uncoated substrata, indicating that substratum properties were at least partly transferred by the adsorbed protein film to the interface with adhering micro-organisms. For S. sanguis 12 and S. mitis BMS, adhesion decreased proportionally with the amounts of sIgA detected by ELISA, but for S. mutans NS such relations with the amounts of sIgA detected on the protein-coated substrata were not found. Thus, for S. mutans NS a specific antibody effect seems to exist in addition to a non-specific protein effect like that observed for S. sanguis 12 and S. mitis BMS.

Amount and surface structure of albumin adsorbed to solid substrata with different wettabilities in a parallel plate flow cell

In this article we studied the adsorption of serum albumin to substrata with a broad range of wettabilities from solutions with protein concentrations between 0.03 and 3.00 mg.mL-1 in a parallel-plate flow cell. Wall shear rates were varied between 20 and 2000 s-1. The amount of albumin adsorbed in a stationary state was always highest on PTFE, the most hydrophobic material employed and decreased with increasing wettability of the substrata. Increasing stationary amounts of adsorbed albumin were observed with increasing wall shear rates at the lowest protein concentration. Inverse observations were made at the highest protein concentration. Transmission electron micrographs of replicas from the albumin-coated substrata showed that proteins were mostly adsorbed in islandlike structures on the hydrophobic substrata. The tendency to form islandlike structures was shear rate- and concentration-dependent and disappeared gradually going to more hydrophilic substrata. On glass, the most hydrophilic material employed, a homogeneous, well distributed, fine knotted, reticulated structure was found. In conclusion, this study demonstrates that both the amount of adsorbed albumin as well as the surface structure of the adsorbed proteins are regulated by the substratum wettability. This observation may well account for the fact that substratum properties can be transferred by an adsorbed protein film to the interface with adhering cells or microorganisms.

Blood plasma proteins on polyurethane and alkylsiloxane plasma-treated polyurethane surfaces. Dynamic approach by stimulus-response technique. Part 2. Evaluation of adsorption data by moment technique

In this study, interactions of blood proteins (i.e. albumin and fibrinogen) with polyurethane biomaterial surfaces were investigated in an in vitro bead column test circuit using a stimulus-response technique. The dynamic sorption process of radiolabelled proteins on the surfaces was followed by detecting the radioactivity at the exit stream of the column, which was the response of a pulse stimulus at the inlet. The mathematical model was described and solved using 'parameter estimation by cybernetic moment technique', and the adsorption rate constants of plasma proteins on different biomaterial surfaces were calculated. By evaluation of the response curves with standard and cybernetic moment techniques, the following results were obtained. Albumin and fibrinogen adsorption is competitive, and the competition is strongly dependent upon the surface characteristics of the biomaterial. Preadsorption or preferential adsorption of albumin decreases the fibrinogen adsorption, and therefore increases the biocompatibility of material surface. Adsorption of blood plasma proteins are irreversible. The moment technique can also be used for the evaluation of stimulus-response data of biological systems, to determine the process parameters.

Bulk, surface, and blood-contacting properties of polyetherurethanes modified with polyethylene oxide

The bulk, surface, and blood-contacting properties of a series of polyether polyurethanes based on polyethylene oxide (PEO) (MW = 1450), polytetramethylene oxide (PTMO) (MW = 1000), and mixed PEO/PTMO soft segments were evaluated. The effect of varying the weight percentage of PEO, and thus the overall polarity of the mixed soft segment phase, was investigated. Two polymer blends prepared from a PTMO-based and a PEO-based polyurethane were also studied. Differential scanning calorimetry (DSC) and dynamic mechanical analysis indicated that the polyurethanes based on either the PEO or the PTMO soft segments are relatively phase mixed. The degree of phase mixing in the polymers increased with increasing weight fraction of PEO. As expected, water absorption and the hydrophilicity of the polymer increased with increasing PEO soft segment content. In vacuum, the PEO-rich polymers have a lower concentration of soft segment at the surface, possibly due to the migration of the polar PEO segments away from the polymer/vacuum interface. The blood-contacting results indicated that the higher PEO-containing polymers were more thrombogenic than the pure PTMO-based polyurethane. A threshold concentration of PEO in the polyurethane appeared to be required before the blood-contacting properties were significantly affected.

Changes in the SDS elutability of fibrinogen adsorbed from plasma to polymers

Baboon fibrinogen was adsorbed from diluted plasma solutions to glass, polyethylene, polystyrene, polydimethylsiloxane, poly(ethylene)terephthalate, and Biomer. Following adsorption, half of the samples were immediately placed in sodium dodecyl sulfate (SDS) solutions to elute the protein, while the others were stored in buffer for up to 5 days and then eluted in SDS. The elution was typically incomplete, but depended on the plasma concentration and the residence time. The elutability was generally lower for fibrinogen adsorbed from more diluted plasma, and substantially lower for samples on which the fibrinogen had resided for 5 days. The non-elutable portion of the protein layer formed rapidly on polystyrene, while on polyethylene elutability was high initially, followed by a gradual decrease.

Adsorption of plasma proteins and platelet adhesion on to polydimethylsiloxane/poly(gamma-benzyl L-glutamate) block copolymer films

A-B-A-type block copolymers of four different compositions were synthesized, in which A and B represent poly(gamma-benzyl L-glutamate) and polydimethylsiloxane segments, respectively. Among the block copolymers and their surface-modified derivatives, those containing 40-70 mol% polydimethylsiloxane and having water contact angles ranging from 50 degrees to 85 degrees were found not to induce conformational change of plasma proteins upon adsorption. It was also observed that the number of adhered platelets and the rate of serotonin release from adhered platelets increased when plasma proteins underwent conformational change upon adsorption. These experimental observations indicate that hydrophobic-hydrophilic block copolymers having a certain composition do not induce conformational change of plasma proteins upon adsorption and do not adhere to and activate platelets, thus leading to a suppression of thrombus formation.

Quantitative surface studies of protein adsorption by infrared spectroscopy. II. Quantification of adsorbed and bulk proteins

Attenuated total reflectance Fourier transform infrared spectra of surface-adsorbed proteins are correlated with concentration measurements determined by 125I-labeled proteins. This paper demonstrates that linear correlations between the intensity of the major bands of proteins and the quantity of proteins can be obtained for human albumin and immunoglobulin G up to surface concentrations of approximately 0.25 microgram/cm2. A poorer correlation was observed for human fibrinogen. A linear correlation was also observed between the concentration in the bulk solution and the major bands of albumin up to a concentration of 60 mg/ml.

Functional mechanisms of polymer-based in vivo reference electrodes

We describe a reference electrode catheter based upon a haemocompatible porous liquid junction of poly(2-hydroxyethyl methacrylate) (pHEMA). The diffusion properties of pHEMA rapidly reach a steady state in a variety of physiological environments. To elucidate the functional mechanisms, the transport of ions through pHEMA membranes was studied in electrolyte solutions, plasma solutions and whole blood. Plasma was shown to enhance ion transport by approximately 10% whereas blood decreased transport rates by 40%. The stability of the reference electrode catheter remained within 1 mV over an 8 h period. The mechanisms of stability lie in those material properties of pHEMA which control diffusion, limit protein adsorption and respond to changes in pH, properties which may result in micromechanical fluctuations and subsequent renewal of the polymer/blood interface.