Stagnant versus dynamic conditions: a comparative adsorption study of blood proteins.

Haemodynamic parameters of flowing blood, such as diffusion, convection, flow and shear rates, are important as they determine the interaction of cells with vessel walls and prosthetic implants in the cardiovascular system. Most of the studies under flow conditions have been performed with platelets or other cells, and less attention has been paid to the effects that these parameters may cause on the adsorption of proteins. For this reason we studied how different shear rates affect the adsorption of human albumin, fibrinogen, total serum proteins, and complement factors 1q and 3c from human serum to silicon surfaces. The most relevant results indicate that during non-flow conditions the amount of adsorbed proteins is always lower than under flow. The different shear rates (225, 915, 1800 and 2700 s(-1)) all gave similar results, indicating that such a parameter is not very critical for single protein deposition. The differences in kinetics of complex protein solutions are conveniently highlighted by use of specific polyclonal antibodies. The difference between non-flow or low shear rate conditions and physiological flow conditions was enhanced for the complement cascade system.

Imaging of the Early Events of Classical Complement Activation Using Antibodies and Atomic Force Microscopy

In the present report we use atomic force microscopy (AFM) combined with antibody techniques to study the lateral distribution of specific serum proteins adsorbed onto flat silicon surfaces precoated with immunoglobulin G (IgG). Null-ellipsometry was used as a complimentary technique to quantify the adsorbed protein layers. After 15 s of incubation in human blood serum a partial monolayer of randomly distributed serum proteins was observed. The following exposure to antibodies to complement factor 1q (anti-C1q) resulted in a development of enlarged protein aggregates and a significant increase in adsorbed mass. Conversely, exposure to antibodies to complement factor 3c (anti-C3c) resulted in only a few randomly distributed protein aggregates and a much smaller increase in adsorbed mass. After 60 s of serum incubation the entire surface was covered with a proteinaceous film with irregular topography. This layer bound large amounts of anti-C3c but showed significantly smaller affinity for anti-C1q. Prolonging the serum incubation to 30 min resulted in an increased thickness and roughness of the protein layer and caused a massive deposition of anti-C3c but no anti-C1q. The results suggests that the transient affinity of anti-C1q, seen on various classically complement activating surfaces, is due to a shielding of the initially adsorbed proteins by subsequently deposited layers of C3. The results also show that qualitative information of the lateral organisation of specific proteins in a heterogeneous mixture can be assessed using AFM in combination with immunological techniques.

Blood protein interactions with titanium surfaces.

Protein adsorption and complement activation were studied on thin evaporated films of titanium (Ti). The surfaces were cleaned in either a radio frequency (RF) plasma unit, or washed sequentially in trichloroethylene, acetone, ethanol, and water. Both methods resulted in hydrophilic surface with low carbon contamination on the outermost oxide (approximately 11-13 at%). In situ ellipsometry suggested that Ti is an intrinsic coagulation activator in vitro, since significant amounts of factor XII (F XII) and high molecular weight kininogen (HMWK) were found on the surfaces after 1 min incubation in heparin plasma. Ellipsometry, performed after serum incubations ranging from 15 s to 30 min showed that the total amount of serum proteins and the deposition of antibodies to complement factor 3c (C3c) increased with serum incubation time. ELISA methods showed increased levels of free iC3b in serum after 10 min incubation of the surfaces, but no detectable amounts of C3 convertase fractions C4d or Bb. Ellipsometric results indicated, however, an increased deposition of antibodies to CIq and IgG on Ti after short serum incubation times. The combined results indicate that Ti-surfaces initially activate complement through the classical pathway. The activation then continues via a positive amplification loop where increased amounts of C3 are deposited on the surfaces via the alternative pathway.

Blood protein interactions with chromium surfaces.

Protein adsorption, contact activation, and complement activation were studied on thin evaporated films of chromium (Cr) in vitro. The surfaces were, prior to the experiments, cleaned in either ethanol and water, or in a basic peroxide solution (RCA standard clean 1, SC-1). Surface spectroscopic studies of the outermost oxides showed a significant reduction of carbon contaminants after washing in SC-1 but also suggested an increase in the oxidation state as compared with the ethanol-washed surfaces. In situ ellipsometry combined with antibody techniques was used to determine protein deposition and antibody binding onto surfaces after incubations in heparin plasma or in normal serum. Incubation times from 1 to 10 min in serum showed increased depositions of serum and antibodies to complement factor 3c (C3c) and was larger on ethanol-washed surfaces than on surfaces washed in SC-1. ELISA methods indicated increased amounts of iC3b in serum for both surfaces, but no presence of C3 convertases (C4d or Bb fractions). A low or transient complement activation via the classical pathway was indicated on ethanol washed Cr, since deposition of secondary antibodies to complement factor Iq (CIq) was observed only after short incubation times in serum. No procoagulant activity of Cr was indicated, since only low amounts of antibodies to factor XII (F XII), prekallikrein (PKK), and high molecular weight kiniogen (HMWK) bound to the surfaces after incubations in heparin plasma. These results were confirmed using a colorimetric assay where the relative amounts of free plasma kallikrein was assessed using a chromogenic substrate, H-D-Pro-Phe-Arg-pNA (S-2302).

Differential surface binding of albumin, immunoglobulin G and fibrinogen.

Protein-protein interactions, as well as the nature of the surface, significantly affect the activity of a specific protein towards a defined surface. Indications are that protein-protein associations may affect antibody detectability, but in some cases this is the result of altered antigenic accessibility rather than physical removal of the molecule. The antibody binding patterns are also quite variable over an entire methyl-silanol wettability gradient on silicon, suggesting that the surface itself is affecting protein-protein and protein-protein-surface associations. Ellipsometric studies were carried out on the gradients which were incubated in single; binary and tertiary physiological concentration solutions of human albumin, immunoglobulin G (IgG) and fibrinogen. The ellipsometric-antibody detectability of the proteins on such surfaces were found to be variable, depending upon the location on the gradient and the order and combination in which the proteins were presented to the surface. Radiolabelled proteins were studied on discrete regions of these gradients. Competitive effects of albumin were found to be inhibitory (negative) with respect to IgG adsorption on hydrophobic surfaces, while enhancing IgG deposition on hydrophilic surfaces (positive). Scanning force microscopy in the so-called tapping mode indicates that proteins, particularly IgG, organize themselves differently with respect to surfaces, depending upon the nature of the surface and the presence of other proteins.

Titanium with different oxides: in vitro studies of protein adsorption and contact activation.

Adsorption of albumin (HSA) and fibrinogen (Fib) from human blood plasma onto titanium surfaces with varying oxide properties was studied with an enzyme-linked immunosorbent assay. The intrinsic activation of blood coagulation (contact activation) was studied in vitro using a kallikrein-sensitive substrate. The sample surfaces were characterized with Fourier transform Raman spectroscopy. Auger electron spectroscopy and atomic force microscopy. Low Fib and high HSA adsorption was observed for all titanium samples except for the radio frequency plasma-treated and water-incubated samples, which adsorbed significantly lower amounts of both. Oxide thickness and carbon contamination showed no influence on protein adsorption or contact activation. Smooth samples with a surface roughness (Rrms) < 1 nm showed some correlation between surface wettability and adsorption of Fib and HSA, whereas rough surfaces (Rrms > 5 nm) did not. To varying degrees, all titanium surfaces indicated activation of the intrinsic pathway of coagulation as determined by their kallikrein formation in plasma.

An in-vitro study of H 2O 2-treated titanium surfaces in contact with blood plasma and a simulated body fluid.

Ellipsometry and antibody techniques were used to investigate plasma protein adsorption onto titanium (Ti) surfaces pretreated in hydrogen peroxide (H 2O 2). Surfaces preincubated for short times in 10 mM or 100 mM H 2O 2 bound relatively large amounts of anti-high molecular weight kininogen (a-HMWK) after immersion in blood plasma. Increasing the preincubation time in H 2O 2 led to an increase in the total amount of bound plasma proteins and a large deposition of anti-fibrinogen (a-Fib). Large amounts of a-HMWK and a-Fib were also deposited onto surfaces washed in trichloroethane, acetone and ethanol, whereas radiofrequency plasma-treated surfaces or surfaces incubated in deionized water bound preferentially a-HMWK after plasma immersion. Auger electron spectroscopy (AES) made on Ti-surfaces preincubated in 10 mM and 100 mM H 2O 2 solutions showed an increased oxide thickness and, after 16 h of immersion in a physiological buffer, an increased amount of calcium on and throughout the oxide. The rate of net oxide growth was larger in 10 mM than in 100 mM H 2O 2.

Antisera binding onto metals immersed in human plasma in vitro.

Titanium (Ti), silicon (Si), silver (Ag), vanadium (V), gold (Au), and chromium (Cr) surfaces were made hydrophilic by radiofrequency plasma treatment. The binding of antifibrinogen (a-fib) and anti-high-molecular-weight kininogen (a-HMWK) after incubation of surfaces in 10% human blood plasma was investigated with ellipsometry. Ti and Si surfaces bound a-HMWK but no detectable amounts of a-fib, whereas the other metals bound both types of antisera. Protein adsorption in a fluid gradient produced under a convex lens indicated that the small deposition of a-fib on Ti and Si surfaces is probably a result of protein displacement. This displacement was not evident at lens positions simulating low plasma concentrations where large amounts of a-fib adsorbed. Surfaces washed sequentially in trichloroethane, acetone, and ethanol produced more hydrophobic surfaces and resulted in slower rates of a-fib displacement. These studies would indicate that some metals form complexes with specific plasma proteins or, alternatively, that different metals may bind equal amounts of plasma proteins which express differing antigenicities toward their surroundings.