Characterization of a self-regulating insulin delivery system

The development of a self-regulating insulin delivery system to modulate insulin release in response to blood glucose levels provides a unique and novel approach for the treatment of diabetes. The system is based on the synthesis of an insulin derivative capable of binding to the saccharide site on Concanavalin A. The insulin derivative is released in proportion to the amount of glucose competing for the binding sites. The focus of this dissertation is the characterization of the components involved in this delivery device. Glycosylated insulin derivatives were synthesized for use in this self-regulating insulin delivery system. Due to the synthetic procedure, heterogenous insulin derivative products were created. These were separated and characterized to determine their degree and site of substitution. The major product was determined to be a B-1 phenylalanine monosubstituted insulin conjugate, and had a biological potency equivalent to native insulin. Permeation of glycosylated insulin and glucose through various polymer membranes was investigated in order to optimize the diffusional processes. A Nucleopore polycarbonate membrane, with 0.1 /µm pores, was found to be the most suitable rate controlling membrane for use in this delivery system. Con A immobilized macrobeads were utilized to prevent the diffusion of Con A from the system. Studies demonstrated that the immobilization of Con A does not affect its binding affinity for either glycosylated insulin or glucose. The feasibility of the self-regulating insulin delivery device was examined in both column and in vitro release experiments. These studies showed that the release of bound glycosylated insulin was sensitive to different glucose concentrations under flow conditions. Mathematical models for both the column and in vitro release experiments were utilized to study these phenomena. In vivo plasma disappearance and liver uptake experiments in dogs revealed that there are no significant differences between the manner in which the body treated glycosylated insulin relative to native bovine insulin. In addition, no immunological response to the glycosylated insulins was observed in a rabbit model.