Oral tolerance is a well-characterized phenomenon in animals and is highly effective when induced as a treatment for experimental autoimmune disease. However, its use as a therapeutic modality for the treatment of autoimmune disease in humans has been disappointing. Much of the rationale for its use in humans is based on the finding that feeding antigen to rodents elicits regulatory T cells in Peyer's patches (PPs) that secrete immunosuppressive cytokines such as transforming growth factor (TGF)-beta. By contrast, human antigen-specific PP T-cell responses, and mucosal T-cell responses in general, are strongly biased towards T helper 1 (Th1) cells, which are pro-inflammatory rather than immunosuppressive. This is caused by the high local levels of interleukin (IL)-12 in PPs.