No abstract

The oxidation of antibody carbohydrate residues by periodate is a common approach used for site-specific antibody modification and immobilization. This study sought to develop a general kinetic model that could be used to describe the effective rate of this oxidation for process control. A detailed analysis of previous data collected for rabbit immunoglobulin G in the presence of excess periodate indicated that the reaction followed a pseudo-first-order mechanism in which two general classes of sites were being oxidized. The first class of sites was oxidized fairly rapidly (i.e., within 15-30 min), while the second class of sites reacted over the course of several hours. From these results, an equation was developed that gave a good fit under a variety of reaction conditions to the production of oxidized sites available for coupling with a hydrazide label. On the basis of this equation, data obtained at several periodate concentrations under the same pH and temperature conditions were used to estimate the apparent rate and equilibrium constants for the oxidation of each class of sites. The values obtained by using this approach could be used not only to predict the effective rate of oxidation at other periodate concentrations but also to provide information on the individual steps involved in the oxidation process.

Several factors can potentially affect the rate of immobilization of proteins onto solid supports, such as those used in affinity-based high-performance liquid chromatography. This study examined several of these factors and their influence on the coupling of periodate-treated rabbit immunoglobulin G antibodies to dihydrazide-activated silica. Items considered included the number of potential coupling sites on the antibodies, the density of activated sites on the support, the relative amount of antibody combined with the support, and the density of the overall reaction slurry. In each case, the rate of change in the solution-phase antibody concentration gave biphasic behavior which could be described by two competing pseudo-first-order reactions. The overall immobilization rate was essentially independent of the density of the support's activated sites (when present at a coverage of 0.1-0.4 micromol/m2) but was strongly influenced by the number of available coupling groups on the antibodies. Increasing the slurry density had no appreciable effect on the immobilization rate, and the reaction rate showed only a small change when using different types of reagents for support activation (e.g., adipic vs oxalic dihydrazide). These results are consistent with a mechanism in which the rate-limiting step during immobilization is the covalent attachment of antibodies to the support and not mass transfer of antibodies to the support's surface.

Figure 1. Schematic of an FIA system for performing the BCA protein assay.