Co-stimulation in gut mucosal immunity. Evidence of a differential regulationof local and systemic responses.

In recent years we have learned that co-stimulatory factors are required as second signals for optimal T cell stimulation. Most types of immune responses are, therefore, dependent on the co-stimulatory signals that are provided by infection or immunization. The gut mucosal immune system is continuously exposed to a large number of antigens, delivered by microorganisms or as food antigens. It is of necessity that the system can distinguish between food antigens and antigens that may be harmful to the body. In healthy individuals there is a delicate balance between tolerance and active immune responses. To what extent co-stimulatory factors influence this balance is incompletely known. In the present thesis I have asked whether co-stimulatory factors affect the induction of mucosal IgA responses and whether these factors impact on the regulatory function of gut CD4+ T cells.A prerequisite for this work was the existence of genetically modified mice that provided me with experimental models that were deficient for the most important co-stimulatory factors known today. Experiments were designed to perform oral or systemic immunizations to compare the influence of a defined co-stimulatory defect on mucosal IgA immunity with that of normal mice. To this end we used various protein antigens given together with cholera toxin (CT), which is a potent mucosal immunogen and adjuvant. I also sought to investigate whether induction of mucosal and systemic IgA immunity are differently regulated. Our results reveal a paradox in the regulation of mucosal IgA immune responses. On the one hand mucosal total IgA production was intact in mice that had a defect in the B7-CD28 co-stimulatory pathway. By contrast, such mice failed completely to respond to an oral immunization. Thus, total gut IgA production must be under the influence of at least two regulatory pathways. Moreover, mice deficient in CD28 demonstrated intact mucosal immune responses to oral immunization, but poor systemic responses, suggesting that mucosal and serum IgA responses are differentially regulated and require different types of co-stimulatory signals. Strong mucosal, but poor systemic IgA responses in CD28-deficient mice suggest that other molecules, redundant with CD28, may govern mucosal but not systemic responses. Mice with a defect in CD19, a co-receptor on the B cells important for B-1 B cell development, had normal gut total IgA levels, but failed to respond to oral immunizations, suggesting that B-1 B cells are dispensable for normal total gut IgA production. Experiments in the CD19-deficient mice revealed a defect in CD40-signalling, which strongly impaired Th2 co-ordinated B cell responses such as IgG1, IgE and gut IgA responses, but it had no effect on Th1-coordinated IgG2a or serum IgA responses. Although unresponsive to oral immunization, CD40-deficient mice confirmed that a substantial production of total gut IgA is CD40-independent. It is probable that TI antigens of microorganisms, resident in the gut lumen, promote these IgA responses.Several important conclusions were drawn. The first is that many of the models exhibit gut IgA responses irrespective of the presence or absence of germinal centers (GC) in the Peyer's patches, suggesting that many IgA precursor B cells are derived from other sites. However, the factors that determine GC formation clearly differ between gut PP and systemic lymph nodes or spleen. Secondly, B-1 B cells appear to play a minor role for the total IgA production. Thirdly, CT is a potent adjuvant which can act independently of CD28 and B7-costimulation, a finding contradicting the results of several previous studies