A computerized protein–protein interaction modeling study of ampicillin antibody specificity in relation to biosensor development

Abstract

Nonspecific interactions between immobilized biomolecules and interfering proteins significantly impede biosensor development and commercialization. Advances in bioinformatics and computer technology have facilitated a greater understanding of biological interactions. We employed two different protein-protein docking programs to simulate the nonspecific interaction between ampicillin antibody and potential interfering proteins (human serum albumin and ovalbumin). To evaluate the contact and probability of association with the active site of the antibody, different amino acid chains from human serum albumin (HSA) and ovalbumin (OVA) were modeled in the simulation. In addition, a well-known specific immune complex, lysozyme and lysozyme antibody, was simulated for comparison. The results demonstrated that the cluster density of nonspecific interactions was smaller than the specific interaction between lysozyme and antibody, and that the dock scores were scattered. However, the active site of ampicillin antibody was prone to nonspecific protein interactions. The strength of interaction was different for specific binding and nonspecific binding. These results provide a platform for detecting the probability of nonspecific interactions and for improving methods of biosensor detection construction with reduced nonspecific adsorption.