Therapeutic proteins provide effective treatments for numerous human diseases and medical conditions, such as diabetes, anemia, and cancer. Utilization of a protein as a therapeutic agent requires preservation of the native protein structure since any degradation of this structure can lead to a reduction in therapeutic efficacy and can cause toxic side effects. This requirement poses a daunting challenge since proteins are only marginally stable and are susceptible to numerous forms of chemical and physical degradation. Coupling the marginal stability of proteins with the deleterious effects of degradation underscores the need for a cogent methodology to use when developing a therapeutic protein product.

The central objective of this dissertation is to develop and apply a more rational approach to protein formulation development that addresses one of the many protein degradation pathways: surface-nucleated protein aggregation. To accomplish this objective, we use recent advances in protein and colloid chemistry, as well as well-established principles in solution thermodynamics, physical chemistry, and reaction kinetics to gain both qualitative and quantitative insight into the factors involved in the aggregation process. Since the nucleation-dependent aggregation process can be either homogeneous or heterogeneous, we account for both types of aggregation in our work.

We investigate homogeneous nucleated aggregation using two insulin analogues, biosynthetic human insulin and lispro insulin, as model proteins. We show that the aggregation of each protein follows a nucleation growth process where native insulin molecules irreversibly assemble to form long, fibril aggregates. Furthermore, we demonstrate how a difference in the propensity of the insulin molecules to reversibly self-associate affects the irreversible aggregation process.

In order to investigate heterogeneous nucleated aggregation, we use the silicone oil found in medical syringes to induce the aggregation of four model proteins; lysozyme, bovine serum albumin, abatacept, and trastuzumab. To quantify the amount of silicon oil-induced protein aggregation with common techniques, we utilize a silicone oil-in-aqueous solution emulsion. Results indicate a constant amount of aggregation over a two week period for all four proteins. The extent of aggregation decreases or increases in the presence of surfactants or sodium chloride, respectively. We demonstrate that silicone oil-induced aggregation is protein independent with the exception of lysozyme, which appears to undergo a bridging flocculation phenomenon.

The remainder of this dissertation involves the continuation of our investigation into heterogeneous nucleated aggregation. Our final objective is to determine the mechanism of silicone oil-induced aggregation using flow cytometry, a tool often used in cell biology but not typically used in the development of therapeutic protein formulations. Results indicate protein adsorption onto the silicone oil surface as the primary aggregation mechanism.