Split-peak determination of protein adsorption kinetics in high-performance affinity chromatography

Equations were derived to describe the unusual chromatographic situation in which part of a sample of pure analyte elutes in the nonretained peak while the remainder is strongly retained. It was shown that this behavior is the result of slow diffusion and/or adsorption of analyte within the column. Using the equations developed, kinetic parameters were measured for the binding of immunoglobulin G to immobilized protein A affinity columns. By comparing protein A columns prepared using different immobilization methods and support materials, it was found that diffusion was rate-limiting in some cases while adsorption was rate-limiting in others. One immobilization method was found to decrease the adsorption rate constant of the immunoglobulin G-protein A system by an order of magnitude vs. that obtained with other methods;In the second study, the same equations were used to optimize analyte adsorption in a dual column affinity chromatographic system. This system consisted of two columns in series, the first containing immobilized anti-albumin antibodies and the other containing protein A. These were used for the analysis of albumin and immunoglobulin G, respectively. In optimizing adsorption, the column size required by the antibody support was limited by its binding capacity, while the protein A support was limited by the rate of immunoglobulin adsorption. Based on this and other information about the supports, a system was designed for the determination of both proteins in serum. This method gave results in good agreement with commercially available methods, while requiring only 2 (mu)L of serum and 6.0 min per cycle. It was shown that both albumin and immunoglobulin G were selectively retained, with little interference from other serum components;The last study examined the effect of nonlinear elution conditions on the relative size of the nonretained peak. Computer simulations were used to study those cases in which either diffusion or adsorption was rate-limiting. The simulation data were then compared to results obtained for the adsorption of hemoglobin on reversed-phase columns and the binding of immunoglobulin G to protein A columns. From the simulations, guidelines were developed for minimizing nonlinear elution effects in both of the cases studied