Date of Award

Summer 8-2017

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Biological Sciences


Jonathan S. Minden


In the last 20 years, proteomic studies have not yielded any FDA-approved protein antigen biomarkers for diseases. Commonly used antigen discovery method called Serological Proteome Analysis (SERPA), is not useful in identifying low abundance proteins. On the contrary, Antibody mediated identification of antigens (AMIDA) enriches low abundance protein targets and we believe could improve the discovery rate of true antigen biomarkers using proteomics. However, AMIDA has not been popular due to technical challenges. A major limitation is the contamination posed by antibodies that are used for the isolation of antigens. Antibodies being in high abundance, mask the signal of protein antigens and obstructs the mass spectrometry identification of antigens during the discovery phase of autoantigen biomarker screening. In an effort to improve the discovery of protein antigens, we present here a solution using a reversible protein capture reagent, called Biotin-CDM. Biotin-CDM can be incorporated into AMIDA in order to remove contaminating antibodies and enrich low abundance protein antigens. We use Biotin-CDM to reversibly tag all potential target proteins in a cell lysate with biotin. The presence of biotin coupled to the target proteins allows for a secondary separation step in which antibodies are washed away from the reversibly biotinylated target proteins by binding them to an Avidin-coupled matrix. The captured target proteins are released from the Avidin matrix by reversing the Biotin-CDM link, thus releasing a pool of target proteins ready for further proteomic analysis compatible with 2DE. Here, we describe the synthesis of Biotin-CDM and optimization of conditions to label proteins without affecting antibody-antigen interactions. We have successfully incorporated Biotin-CDM in AMIDA to identify antigens targeted by antibodies from Rheumatoid Arthritis (RA) and Interstitial lung disease associated with RA (RA-ILD) patients. Genetic predisposition and cigarette smoking have been linked to a post-translational modification called “citrullination/deimination” that is involved in generating antigens which trigger the formation of neo-epitopes (autoantibodies) in Rheumatoid Arthritis. To understand the pathogenesis of RA-ILD, we screened in vitro deiminated antigens that may be preferentially targeted by antibodies from RA vs RA-ILD patients. We identified several antigens including catalase and cAMP-specific 3',5'-cyclic phosphodiesterase 4D isoform PDE4D5 that immunoprecipitated with antibodies from RA & RA-ILD patients. Surprisingly, the same proteins immunoprecipitated in the treated and deiminated samples. The major changes were in the post-translational modifications (PTM) of the immunoprecipitated proteins. Further exploration into the PTM preference by RA/RA-ILD patient antibodies will need to be done in the future. Finally, we have performed comparative proteomics study on mitochondrial proteomes from Huntington’s disease cell line model. Here we show the benefits of using antibodies for the isolation of mitochondria and the critical importance of sample preparation for comparative proteomics.