Anatomical and Cellular Basis of Immunity and Tolerance in the Intestine

INTRODUCTION The intestine encounters more antigen than any other part of the body and so contains the largest proportion of the immune system. This sophisticated immune apparatus is necessary to generate strong protective immunity against pathogenic infections, but much of the intestinal antigen load consists of harmless materials such as food proteins and commensal bacteria. It would be wasteful to direct active immune responses against these antigens and in fact, such responses cause intestinal disorders such as coeliac disease and Crohn’s disease (1). For these reasons, the default response to harmless antigens in the gut is the induction of immunologic hyporesponsiveness. In addition to its physiological role, this phenomenon of oral tolerance can be exploited to deliver immunotherapy against autoimmune and inflammatory diseases (1,2) and it is also an important obstacle to the development of oral vaccines. ANTIGEN UPTAKE IN THE INTESTINE The conventional view of the mucosal immune system is that immunogenic forms of antigen such as invasive bacteria, viruses or particles are taken up preferentially by M cells in the specialised epithelium overlying Peyer’s patches (PP), before being passed on to professional antigen presenting cell (APC) in the underlying lymphoid tissues. These APC then present the antigen to specific CD4† T cells, either in the thymus dependent areas (TDA) of the PP itself, or after migrating to the MLN. T cells primed in this way are induced to express the α4β7 integrin and the CCR9 chemokine receptor, molecules which mean that the differentiated T cells recirculate specifically from the bloodstream into the intestinal mucosa (3). This pathway is attractive because it can explain the anatomic compartmentalisation of the mucosal immune response. There are also unusual populations of dendritic cells (DC) in PP, some of which have been shown to produce IL10 and polarise naïve CD4† T cells towards a Th2 or regulatory phenotype (4), which are often considered to be the default pathway of intestinal T cell responses. Antigen uptake and presentation restricted to the PP cannot account for all aspects of mucosal tolerance and immunity, especially the fact that hyporesponsiveness of systemic immunity is characteristic of oral tolerance. There is also little direct evidence that M cells play a central role in the uptake of non-particulate antigens such as proteins. Oral tolerance can be induced normally in μMT B cell KO mice that lack M cells and PP (5). Studies using “PP null” mice generated by genetic or induced KO of lymphotoxin β receptor (LTβR) mediated signalling, have also produced highly conflicting results on whether PP are essential for orally administered antigens to tolerise or prime systemic immunity (reviewed in 3). Thus there must be other anatomic pathways for the induction of mucosal tolerance and immunity. These include uptake of antigen via the villus epithelium, with loading of APC in the mucosal lamina propria, or with local presentation of antigen by enterocytes to CD4† T cells in the lamina propria. In addition, there may be blood-borne dissemination of native antigen to local and peripheral lymphoid tissues, with direct loading of APC and presentation to CD4+ T cells in these tissues. ANTIGEN UPTAKE ACROSS VILLUS EPITHELIUM The surface area of the conventional villus epithelium is much greater than that of the PP and in preliminary studies, we have found that a much higher proportion of orally administered soluble proteins associates with APC in the LP than in the PP. Several types of APC, including DC, are abundant in the underlying lamina propria and it has been suggested that some of these DC may be able to enter the epithelium to sample antigens in the lumen (6). LP DC are therefore in an ideal position to sample local antigens and to begin to analyse their role in more detail, we have purified antigen loaded DC from the LP of mice fed high, tolerogenic doses of OVA. Our findings indicate that in vivo loaded LP DC can present the antigen to CD4+ T cells in vitro and in vivo, but interestingly, these DC may induce tolerance when transferred into naïve recipients. LP DC also appear to produce immunoregulatory cytokines such as IL10 and type 1 IFN, rather than pro-inflammatory mediators such as IL12. DC from other tissues do not share these properties, suggesting that LP DC may play a central and unique role in the induction of oral tolerance. As naïve CD4+ T cells are rare in the lamina propria and LP CD4+ T cells do not migrate out of the gut (3), it seems unlikely that local presentation of antigen by DC in the LP could initiate systemic tolerance or immunity. This probably requires loaded DC to migrate from the LP to the draining MLN before meeting naïve CD4+ T cells here. Nevertheless, local APC could be involved in maintaining the survival and activity of already primed regulatory or effector T cells in the LP and these ideas need to be addressed directly. INFLUENCE OF DENDRIC CELLS ON THE INDUCTION OF TOLERANCE VS IMMUNITY DC are central to the induction of both tolerance and immunity to oral antigens. The intestine and its lymphoid organs contain large numbers of DC, including a number of unique subsets (3,4). Many are in ideal positions to take up antigen directly from the lumen and the lymph draining the gut contains DC that can be loaded with antigen by feeding (7). Expansion of DC numbers with the cytokine flt3 ligand (flt3L) enhances the susceptibility of mice to the induction of oral tolerance after feeding soluble proteins, but it can also increase the priming effects when antigen is given together with an adjuvant (8,9). Therefore under physiological conditions, DC from intestinal lymphoid tissues may be in a “quiescent” state, with low levels of costimulatory molecules and producing cytokines such as IL10. Nevertheless, they can be activated by inflammatory stimuli. Together these findings suggest that it may be feasible to manipulate mucosal immunity and/or tolerance via the rational manipulation of DC by altering the nature of the antigen.