Prothrombin adsorption

Prothrombin sorption is reversible and calcium-dependent. The prothrombin association constant K is dependent on the surface concentration of protein and on the composition of the phospholipid bilayers, indicating interacting binding sites. The initial rate of prothrombin adsorption is transport limited in all conditions studied. Values of the sorption rate constants k and are dependent on the surface concentration. The rate of adsorption decreases for higher surface concentration and the intrinsic values of... 

Prothrombin adsorption remains reversible on pure phosphatidylserine (PS) bilayers and on a mixture of 80% PS and 20% phosphatidylcholine (PC). For PS/PC mixtures with less than 80% PS the initial reversible prothrombin adsorption is followed by a slow second surface reaction which causes irreversible adsorption. A similar slow surface reaction is seen for fibrinogen on 100% PS. Elimination of calcium after adsorption of fibrinogen gives a fast desorption of part of the adsorbed layer, possibly due to increased negative charge of the fibrinogen molecules. 

Figure 5. The intrinsic adsorption rate function, as a function of surface concentration, for prothrombin adsorption on a DOPS doublelayer.

By preparing planar lipid monolayers or bilayers on hydrophobically derivatized or native hydrophilic glass, respectively, the adsorption equilibrium constants of a blood coagulation cascade protein, prothrombin, have been examined by TIRF on a surface that more closely models actual cell surfaces and is amenable to alterations of surface charge. It was found that membranes of phosphatidylcholine (PC) that contain some phosphatidyl-serine (PS) bind prothrombin more strongly than pure PC membranes/83... 

It appears from these data that adsorption to hydrophilic surfaces would be a viable strategy. Other research produced a similar conclusion. Waugh et al. showed that prothrombin (another member of the coagulation cascade) binds more strongly to polymethyl methacrylate than glass, ("hating showed that plasma proteins bind more strongly to polyvinyl chloride than a hydrophilic dialysis membrane. 

Morrissey 53) used transmission infrared spectroscopy to study protein adsorption onto silica particles in a heavy water (DzO) buffer. By observing the shift in the amide I absorption band, he could deduce the fraction of protein carbonyl groups involved in bonding to the silica surface. He found that bovine IgG had a bound fraction of 0.20 at low bulk solution concentrations, but only about 0.02 at high solution concentrations. However, neither prothrombin nor bovine serum albumin exhibited a change in bound fraction with concentration. Parallel experiments with flat silica plates using ellipsometry showed that the IgG-adsorbed layers had an optical thickness of 140 A and a surface concentration of 1.7 mg/m2 at low bulk solution concentration — in concentrated solutions the surface amount was 3.4 mg/m2 with a thickness of 320 A (Fig. 17). 

There are several possible arrangements tolerating the presence of particles during adsorption of proteins to particulate matrices. Batch adsorption in stirred tanks is performed by contacting adsorbent particles with a cell containing suspension. After protein capture the adsorbent is separated from the broth and the protein of interest can be eluted. This procedure has been described for the isolation of antibiotics [12], the purification of ot-amylase from B. amylo-liquefaciens broth [13], and the isolation of the prothrombin complex from... 

Preprothrombin can be determined immunologically, either using antiprothrombin antibodies, after adsorption of the y -carboxylated protein onto barium carbonate or using anti-preprothrombin antibodies that do not cross-react with prothrombin. Circulating concentrations of preprothrombin in vitamin K deficiency are of the order of 150 to 1,500 nmol per L. If elevated preprothrombin is because of vitamin K deficiency, then it will fall on administration of the vitamin, whereas if it is the result of liver disease, then vitamin K supplements will have no effect. 

Figure 4. Adsorption of prothrombin from a 60 mg % solution onto polymer surfaces...

The studies on the mode of interaction of prothrombin with phospholipid monolayers, using complementary methods of surface measurement are reviewed. They were investigated at air-water and Hg-water interfaces respectively by radioactivity and electrochemistry. A process more complex than a simple adsorption could be detected. Indeed, the variation of the differential capacity of a mercury electrode in direct contact with phospholipid monolayer, induced by the interaction with prothrombin could be interpreted as a model of its penetration into the layer this was confirmed by the study of the dynamic properties of the direct adsorption of this protein at the e-lectrode, followed in part by the reduction of S-S bridges at the electrode. It could be also concluded that prothrombin resists complete unfolding at these interfaces. 

As an example of a membrane model, phospholipid monolayers with negative charge of different density were used. It had already been found ( ) and discussed O) that the physical and biological behavior of phospholipid monolayers at air-water interfaces and of suspensions of liposomes are comparable if the monolayer is in a condensed state. Two complementary methods of surface measurements (using radioactivity and electrochemical measurements), were used to investigate the adsorption and the dynamic properties of the adsorbed prothrombin on the phospholipid monolayers. Two different interfaces, air-water and mercury-water, were examined. In this review, the behavior of prothrombin at these interfaces, in the presence of phospholipid monolayers, is presented as compared with its behavior in the absence of phospholipids. An excess of lipid of different compositions of phos-phatidylserine (PS) and phosphatidylcholine (PC) was spread over an aqueous phase so as to form a condensed monolayer, then the proteins were inject underneath the monolayer in the presence or in the absence of Ca. The adsorption occurs in situ and under static conditions. The excess of lipid ensured a fully compressed monolayer in equilibrium with the collapsed excess lipid layers. The contribution of this excess of lipid to protein adsorption was negligible and there was no effect at all on the electrode measurements. 

The adsorption of prothrombin onto the lipid monolayer was followed directly by counting the surface radioactivity of the H labelled protein using a gas-flow counter equipped with an ultrathin window as described elsewhere ( ). By calibrating the counter as previously described ( ), it is possible to determine the surface concentration of the radioactive protein, T. ... 

Figure 1. Time dependence of adsorption of 5 yg/ml of prothrombin, at the air-water interface in the presence of a phosphati-...

At the pure air-water interface (Figure 1), the initial rate of adsorption of prothrombin is proportional to the square root of the time, as can be calculated from the curves, indicating that the process, at this interface, is diffusion-controlled, as was observed for native DNA ( ). In the presence of phospholipid monolayers the adsorption process is slower a more complex process must take place at the surface of the membrane, in addition to the conformational change of prothrombin which occurs in the presence of Ca as was found by Nelsestuen (9). It can be the penetration of prothrombin into the lipid layer, as will be described below. 

Direct adsorption of Prothrombin at the Mercury-water Interface. The adsorption rate of prothrombin on a hanging mercury drop electrode (HMDE) was studied by measuring the decrease of the differential capacity with time of contact of the mercury drop with the solution, at a fixed potential, - 0.5 V, in parallel with the increase of the areas of the voltametric peaks corresponding to the reduction of some S-S bonds of the adsorbed molecules (14). The representation on... 

From the kinetics of adsorption, the surface coverage could be obtained. In the case of prothrombin, the number of molecules adsorbed on the mercury surface, T - /2 > could be evaluated from the linear dependence of the capacily on t / only at short times (< 50s), when the diffusion layer thickness (Dt) is still smaller than the thickness of the unstirred layer. In this region the concentration is... 

Figure 4. Surface concentration of electroactive disulfide bonds from the 2nd reduction (i, 1st oxidation sweep , as a function of the decrease of the capacitance of the HMDE in contact with a prothrombin solution at 4.5 yg/ml. Potential of adsorption -0.5V...

With other biological macromolecules (19, 20), the number of adsorbed molecules was usually calculated from the linear dependence of the capacitance on t / using Equation 1, over the whole range of adsorption. The surface concentration of hormones could also be inferred directly from the calculated number of charges transferred between the electrode and an electroactive group, like S-S, of the adsorbed molecules, each one containing only one S-S (21, 22). This method could not be used for proteins, where only part of the S-S are available for the electrode reaction,as seen for prothrombin but in this case of proteins, the method of exploitation of the data presented above is very useful and quite new. 

The prothrombin molecule does not unfold on the different interfaces. In the presence of Ca, some changes of the position of the protein relative to the lipid layer could be detected. Ca also induces stabilization of the globular structure, as measured at a bare mercury electrode. Furthermore, a dynamic picture of the growth of the adsorbed prothrombin layer on mercury electrode was presented and shows 2 distinct adsorption states. 

It was confirmed that Ca increases the adsorption, but it was found that even in the absence of Ca , prothrombin interacts with phospholipids. Consequently, besides the electrostatic interactions, some Ca - independent interactions, which might be hydrophobic, are also involved. The observation that prothrombin and some of its fragments penetrated the layer is in accordance with this idea. This implies that fragments other than Fragment 1 might be involved in the interaction. It must be noticed, that penetration could not be obtained by the less sensitive technique of surface pressure measurements ( ). 

In 1983 we were able to validate two exact formulae, based on the Lorenz-Lorentz equations, allowing the calculation of the mass of the adsorbed layer from the refractive index and thickness. This experimental validation was performed by measuring stacked multilayers of known mass of phosphatidylserine and by the adsorption of radiolabeled albumin and prothrombin on these multilayers (9 ). 

The unstirred layer adsorption model can be generalized by the introduction of surface concentration dependent sorption rate constants k and This subject is currently being studied as well as the existence of a second, irreversible, surface reaction following reversible initial adsorption for fibrinogen and prothrombin on a 60% DOPS/40% DOPC mixture. 

Figure 3. Adsorption and desorption of prothrombin on a DOPS doublelayer. The adsorbed mass is calculated from the changes in the refractive index and thickness, which in their turn are calculated from the changes in polarizer and analyzer readings (see text) ...

In a previous paper (J ) it was shown that the adsorption of prothrombin on a double layer of DOPS in Tris-HCL buffer is transport-limited for low surface concentrations. In Figure 5 this adsorption is analyzed for higher surface concentrations. At a surface concentration of 0.1 yg/cm the adsorption rate drops below the diffusion... 

---