Previous work by our research group showed that PR3-ANCA patients had an antibody response to a recombinant complementary-PR3 protein encoded by the antisense strand of the PR3 mRNA. To follow up on this work, we sought to determine whether the patients also had a T cell response to this recombinant complementary-PR3 protein and whether a protein reactive with those antibodies could be identified in vivo. Chapter 2 of the dissertation describes the identification of CD4+ TH1 cells that proliferate in response to a complementary-PR3 peptide. This proliferation was seen by both a CFSE assay as well as by interferon-γ production in an ELISPOT assay. Those patients who had a T cell response to complementary-PR3 peptide also had antibodies to the complementary-PR3 protein. We next sought to determine if complementary-PR3 proteins could be identified from patient plasmapheresis material. Chapter 3 of this dissertation describes the identification of two complementary-PR3 proteins, human plasminogen and Protein F, a protein from pseudomonas. These proteins reacted with an antibody raised to a peptide encoded by the antisense RNA of the PR3 gene. As complementary proteins are known to interact, plasminogen was shown to be a substrate of PR3, indicative of interaction between the two proteins. Lastly, the anti-complementary PR3 antibodies also bound to normal human leukocytes, cells that are known to bind plasminogen on their surface. Chapter 4 describes the identification of anti-plasminogen autoantibodies in PR3-ANCA positive patients. These antibodies were purified using a complementary-PR3 peptide column, indicating that the anti-cPR3 and anti-plasminogen antibodies are the same. The anti-plasminogen antibodies bound a surface-exposed loop on plasminogen's catalytic domain. Two in vitro assays confirmed the antibodies affect on plasminogen activity. Serological screening of sera indicated that the anti-plasminogen autoantibodies were more prevalent in those PR3-ANCA patients with a clinical history of venous thrombotic events. By designing an experimental approach that considered protein complementarity, a previously unknown autoantigen and its pathogenic autoantibodies were identified. Consideration of complementary proteins can be used to discover other, and perhaps proximal, autoantigens and autoantibodies in other autoimmune diseases.