Surface exclusion and molecular mobility may explain Vroman effects in protein adsorption

Surface-dependent transitions during self-assembly of phospholipid membranes on mica, silica, and glass

Formation of supported membranes by exposure of solid surfaces to phospholipid vesicles is a much-used technique in membrane research. Freshly cleaved mica, because of its superior flatness, is a preferred support, and we used ellipsometry to study membrane formation kinetics on mica. Neutral dioleoyl-phosphatidylcholine (DOPC) and negatively charged dioleoyl-phosphatidylserine/dioleoyl-phosphatidylcholine (20% DOPS/80% DOPC) vesicles were prepared by sonication. Results were compared with membrane formation on silica and glass, and the influence of stirring, buffer, and calcium was assessed. Without calcium, DOPC vesicles had a low affinity (Kd approximately 30 microM) for mica, and DOPS/DOPC vesicles hardly adsorbed. Addition of calcium promptly caused condensation of the adhering vesicles, with either loss of excess lipid or rapid additional lipid adsorption up to full surface coverage. Vesicle-mica interactions dominate the adsorption process, but vesicle-vesicle interactions also seem to be required for the condensation process. Membranes on mica proved unstable in Tris-HCl buffer. For glass, transport-limited adsorption of DOPC and DOPS/DOPC vesicles with immediate condensation into bilayers was observed, with and without calcium. For silica, vesicle adsorption was also rapid, even in the absence of calcium, but the transition to condensed layers required a critical surface coverage of about 50% of bilayer mass, indicating vesicle-vesicle interaction. For all three surfaces, additional adsorption of DOPC (but not DOPS/DOPC) vesicles to condensed membranes was observed. DOPC membranes on mica were rapidly degraded by phospholipase A2 (PLA2), which pleads against the role of membrane defects as initial PLA2 targets. During degradation, layer thickness remained unchanged while layer density decreased, in accordance with recent atomic force microscopy measurements of gel-phase phospholipid degradation by PLA2.

Production of thrombin as a probe for mixing of phospholipids in membranes on solid supports

Phospholipid-covered solid supports have been used successfully as model membranes in studies on blood coagulation and other research fields. In order to produce such membranes, simple exposure of the support to suspensions of phospholipid vesicles was recently introduced, but questions have remained about the process of vesicle adherence to the surface and the physico-chemical properties of the resulting membranes. Using a new technique, mixing of phospholipids in such membranes was demonstrated. A rotating, hydrophilic, silicon disc was exposed in a two-step procedure to vesicles prepared from mixtures of dioleoylphosphatidylserine (DOPS) and dioleoylphosphatidylcholine (DOPC). Factor Xa, factor Va and prothrombin were added and the transport-limited production rate of thrombin was measured. For low surface coverage with 40% DOPS/60% DOPC, a much higher conversion rate was found if, prior to addition of coagulation factors, excess DOPC vesicles were added to fill up vacant surface area. It is concluded that DOPS is spread over the entire surface and that confluent bilayers are formed. The presented technique may also be used to measure lateral diffusion constants.

Monitoring of unbound protein in vesicle suspensions with off-null ellipsometry

In studies on the binding of proteins to small unilamellar phospholipid vesicles (SUV), the concentration of unbound protein usually remains unknown, because the vesicles cannot be separated from the bulk solution. In the present study, this limitation was overcome by addition of a supported planar phospholipid bilayer to the cuvette containing a vesicle suspension. Ellipsometric measurement of the protein adsorption velocities on this bilayer allowed determination of the concentrations of unbound protein. At high protein concentrations the adsorption is rapidly completed and the usual null-ellipsometry is too slow to obtain well-defined initial adsorption rates. Therefore, an off-null technique was developed, allowing measurement of the adsorbed protein mass at time intervals of 20 ms. Binding of prothrombin and coagulation factor Xa was measured in SUV suspensions prepared from a 20% dioleoylphosphatidylserine (DOPS) and 80% dioleoylphosphatidylcholine (DOPC) phospholipid mixture. For prothrombin, a dissociation constant Kd = 140 +/- 27 nM (mean +/- S.E.) and maximal surface concentration gamma max = (8.9 +/- 0.8) x 10(-3) mole of protein per mole of lipid, were obtained. For factor Xa, these values were Kd = 49.6 +/- 6.3 nM and gamma max = (23.0 +/- 1.4) x 10(-3) mole of protein per mole of lipid. These binding parameters are similar to those obtained earlier for planar bilayers. Apparently, the binding of factor Xa and prothrombin is not dependent on surface curvature.