Demonstration of an upper limit to the range of association rate constants amenable to study by biosensor technology based on surface plasmon resonance.

Numerical simulation of BIAcore sensorgrams has highlighted the need for concern about an assumption, inherent in current determinations of rate constants for macromolecular interactions, that the concentration of solute in the flowing phase remains constant at its injected value. This assumption is shown to be valid for systems with effective association rate constants equal to or less than 10 M(-1) values characteristic of antibody interactions with protein antigens. However, the assumption loses validity when the effective association rate constant is raised to 10 M' . The basic correctness of the latter prediction is verified by an experimental study of the interaction between soybean trypsin inhibitor and immobilized -trypsin, a system with comparable reaction kinetics.