Autoimmune disorders are a group of immune-mediated diseases which arise in an individual when their immune system inappropriately recognizes self-components as non-self components. Collectively, autoimmune diseases are estimated to affect 5-8% of the human population and world-wide prevalence is increasing for unknown reasons. Environmental factors are hypothesized to play a pathogenic role in the majority of autoimmune disorders but current tools are ill-suited for studying and/or discovering antigenic players in autoimmunity. To address this problem, cell-surface display methodologies were developed to understand relationships between antibody recognition and environmental pathogens in autoimmune diseases. A methodology for serum antibody epitope mapping using antigen fragment libraries was devised and implemented to elucidate autoimmune disease-specific epitopes present within a common environmental antigen.

Specifically, the findings from this study strengthen the evidence linking Epstein-Barr Nuclear Antigen 1 (EBNA1) epitopes to Systemic Lupus Erythematosus and Multiple Sclerosis, but not Rheumatoid Arthritis, and suggest molecular mimicry between portions of EBNA1 and self-proteins may play a role in propagating these autoimmune diseases. This application serves as a new approach to examining autoimmune disease associations with ubiquitous environmental pathogens, which has been a long-standing challenge. To enable database-driven antigen discovery, a search strategy was developed to identify and expand linear peptide epitopes recognized by Celiac Disease patients' antibodies using randomized peptide libraries. Expansion of identified linear consensus epitopes possessed information content sufficient to unambiguously identify immunodominant epitopes in wheat, barley, and rye cereal grain proteins.

Bacterial clones expressing the expanded epitope discriminated celiac and non-celiac patients (n = 78) with exceptional accuracy in a one-way blind test, yielding 100% sensitivity and 98% specificity. Since sequentially expanded epitope discovery (SEED) does not require a priori knowledge of the environmental antigens or organisms involved in pathology, it may be broadly useful for de novo discovery of environmental antigens that give rise to pathological immune responses.