Autoimmune diseases affect between five and eight percent of the United States population and their pathogenesis is often unknown. Autoimmune diseases can be modulated by pregnancy in one of four ways: symptoms can improve, be unaffected, worsen, or present in relation to pregnancy. Several theories have been proposed to explain the effect of pregnancy on autoimmunity, however, most of these theories are associated with tolerance or immunosuppression and do not account for autoimmune diseases that worsen or present in relation to pregnancy. Part one of this dissertation focuses on an autoimmune disease, autoimmune hypophysitis (AH), that presents in relation to pregnancy. The mechanisms for this association are unclear but we hypothesized that placental proteins may play a role.

By using a mouse model of experimental autoimmune hypophysitis (EAH), our studies demonstrate that emulsified placental proteins uniquely suppress cell and humoral mediated autoimmune responses when co-localized with autoantigens. Conversely, non-emulsified placental proteins enhance EAH, possibly replicating syncytiotrophoblast shedding and representing a potential mechanism by which pregnancy brings about AH. We further demonstrate that placenta-derived immunoglobulin and the TNF superfamily, but not TNF-α, contribute to the suppression of EAH.

Part two of this dissertation focuses on another autoimmune disease, Sjögren syndrome (SS). Although there are several mouse models of SS, none of them fully recapitulate the human disease. By using mice that over express interleukin-12, a cytokine that is elevated in SS, we show that interleukin-12 transgenic mice develop a lung and salivary phenotype similar to SS and thus represent a new model of the disease. Furthermore, we demonstrate that natural killer cells play a dual role in the pathogenesis of SS. By crossing interleukin-12 transgenic mice to interleukin-l5^-/- mice, which lack natural killer cells, we show that natural killer cells induce the lung phenotype but prevent the salivary phenotype.

In conclusion, our studies have enriched our understanding of the pathogenesis of AH and SS. Furthermore, our findings that placenta-derived immunoglobulin may suppress disease in a model of AH and that natural killer cells influence the initiation of disease in a model of SS could eventually lead to therapies for these diseases.