A general method for the assay of binders. Application to the potentiometric determination of albumin in human serum, plasma and whole blood.

From the generalized theory of binding (stepwise equilibrium model, Scatchard model) it can be derived that in a solution containing the binder and a specific ligand, the bound ligand concentration is linearly related to the total binder concentration only if the free ligand concentration is constant. We designed a general method which is based on that principle and we have shown its validity for the albumin assay in serum, plasma and whole blood. We have tested the ligands bromcresol purple and picrate and used ligand-ion selective electrodes to monitor free ligand concentration in a homogeneous solution. The method has excellent linearity. We chose picrate (PIC) as ligand, because of better specificity. Comparison with the bromcresol purple (BCP) and bromcresol green (BCG) spectrophotometric procedures for 93 samples gave the regression equations: y(PIC) = 0.98x(BCP) + 0.76 (g/L, r = 0.979) and y(PIC) = 0.95x(BCG) - 1.3 (r = 0.947), respectively. Within-day and day-to-day precision was up to 1.5% and 2.3%, respectively. Results for albumin in plasma, when transformed to expected whole blood values, were in close agreement with those obtained by direct analysis of whole blood (in the presence of erythrocytes).