Modeling of Insulin Antigen-Antibody Interactions

Thomas, James

Abstract

The goal of this Visiting Scientist application is to extend studies on the human immune response to insulin by using the powerful tool of computer graphic modeling of receptor-ligand interactions for studying the structure of the Fv of an anti- insulin monoclonal antibody (mab) and its interaction with the insulin epitope. Amino acid sequences of human and murine anti- insulin mab, human anti-insulin idiotopes and a murine anti-insulin receptor mab have been determined from nucleotide sequences of the immunoglobulin genes. This project is designed to provide Dr. Nell, as Designated Visiting Researcher, with the capabilities of doing computerized modeling of C alpha backbones of polypeptides, performing molecular dynamics simulations and energy minimization of the structures modeled, predicting amino acid side chain configurations, and graphics analyses of these data for predicting tertiary and quaternary structures of the antigen binding site. This can be used with the existing models of insulin (from X-ray diffraction at 1.5 A resolution) to predict the epitope recognized by the Fv. New information will be gained regarding the structural basis of diversity of human anti-insulin immunity. Knowledge of these techniques will allow Dr. Thomas, laboratory to use the molecular biology data to examine the structural basis for insulin autoimmunity that is part of the autoimmune prodrome of Type I diabetes mellitus. One long range potential of this project is to use predicted models to alter key amino acids on the insulin molecule to decrease immunogenicity and antigenicity of therapeutic human insulin preparations, and potentially for immunoprevention during the autoimmune pathogenesis of Type I diabetes.