Date of Award

Spring 1-1-2013

Document Type


Degree Name

Doctor of Philosophy (PhD)


Chemical & Biochemical Engineering

First Advisor

Theodore W. Randolph

Second Advisor

John F. Carpenter

Third Advisor

Raul Torres

Fourth Advisor

Stephany Bryant


The immunogenic potential of therapeutic proteins has been known for over half a century. However, there is still a debate about the factors that impact immunogenicity. Among the proposed factors, aggregation of protein has been widely accepted to have a critical role on eliciting immune responses. When immune responses are stimulated, anti-drug antibodies (ADAs) may be released in patients, thereby potentially interfering with the safety and efficacy of drugs. The main objective of this dissertation was to show the adjuvant effect of adsorbed-protein-on-particles (heterogeneous aggregates) on immunogenicity. We evaluated the potency of adsorbed murine monoclonal antibody (as an antigen drug) on microparticles (glass or silicone oil) in eliciting immune responses in mice. The immune responses of mice against receiving the antigen drug adsorbed-on-particles were compared using two mouse models. In addition, the effect of dosing levels on ADA response was monitored. To measure the increase in ADA levels after the antigen drug was administered, an Enzyme-linked Immunosorbent Assay (ELISA) was performed, and the pharmacokinetic profile of the model antigen was monitored. The model antigen was a murine monoclonal antibody (mAb1) which was generated in the C57BL/6J mouse strain. In this work, mAb1 was aggregated by various pathways: chemically altered by exposure to UV light or physically adsorbed onto industrially relevant particles. The aggregates were then tested in two mouse models: C57BL/6J, which was the host strain for mAb1 generation, and BALB/c, which does not have the gene to produce IgG2c (model antigen). Certain types of aggregates, mAb1 exposed to UV light or adsorbed onto particles, were found to be more immunogenic than the native form of the model antigen. In addition, immunogenicity was found to be related to the mouse strains that were tested in this study. Different dosing levels of mAb1 were tested on both mouse strains while ADA levels were monitored along with the pharmacokinetic (PK) profile of the drug. In the present study, the stock solution of mAb1 was centrifuged before preparation of the samples for injection to reduce levels of subvisible particles. The results of the aforementioned study revealed that the lower dose had the same or higher potency in eliciting an immune response as the higher dose. However, the level of the response and the number of the responders depended on the strain that was tested. The antibody responses in BALB/c mice were stronger likely due to the foreignness of mAb1 in this mouse strain. In addition, the results of the PK assay showed that the level of mAb1 in BALB/c mice decreased very quickly, most likely as the result of anti-mAb1 antibodies circulating in the system. The remainder of this dissertation investigates the effect of acid- and pressure-induced dissociation on anti-drug/drug immune complexes prior to ELISA assay. Monoclonal antibodies are known to have a long circulating half-life (2-3 weeks). This characteristic increases the chance for the formation of immune complexes between mAb1 and anti-mAb1 antibodies. These complexes may interfere with the ELISA assay. Due to this interference, a low ADA level or a false negative result may be reported. To reduce the interference from the presence of immune complexes, an acid-dissociation step was added to the standard ELISA assay. To test the effects of a low pH environment on the breaking of immune complexes, a complementary study was conducted. In the complementary study, BALB/c mice were introduced to ovalbumin (OVA) as an antigen drug. The injecting/bleeding schedule and ELISA protocol were the same as that performed in the main study. Acid-induced dissociation was found to be a useful step in breaking immune complexes formed between OVA and anti-OVA antibodies, and a higher level of ADAs could be detected. Therefore, an acid-dissociation step was included in the conventional ELISA in the main study.