Introduction: The mucosal immune system is required to respond to intestinal pathogens, yet tolerate the luxuriant commensal flora of the lower small intestine and the colon. In addition to the epithelial cell barrier and IgA immune exclusion, regulation of intestinal T cell responses to commensal bacterial antigens is suggested by spontaneous chronic intestinal inflammation dependent on the presence of the commensal intestinal flora occurs in immune targetted lesions including α/β T cell receptor, IL-2 or IL-10 knockout mice. We have studied regulation by human intestinal CD4+ cells to responses to purified commensal Escherichia coli proteins.Methods: Proliferation of control human peripheral blood (PBL, 5 10 4 /well, 140 hours, 37°C, 5% CO 2 ) and/or intestinal mucosal lymphocytes (IML, 5 10 4 ) responding to autologous antigen presenting cells (APC, 3 10 5 ) prepulsed with purified (cytoplasmic or inner membrane) commensal E. coli proteins where external antigen was washed off prior to irradiation and culture.Results: Whilst PBL (SI3.4-11, n = 33) but not IML (SI < 1.5, n = 33) respond to these purified commensal E. coli proteins, both respond to stimulation with both PHA and IL-2. CD4+ depleted PBL did not respond. Where both PBL and IML were cultured with commensal E. coli prepulsed APC, the PBL response was prevented (SI < 2, n = 33), however this inhibition was reversed when the CD4+ subset is first subtracted from the IML using monoclonal antibody-coated magnetic beads. Moreover there was no effect on the PBL response to non-intestinal recall antigens in the presence of TML (eg tetanus toxoid n = 14) indicating that this downregulation from CD4+ IML is specific to intestinal luminal antigens. The inhibitory effect of IML on PBL proliferation to commensal bacterial antigens was shown to be independent of cell contact using transwells to separate PBL and IML in culture by a semipermeable membrane; however prepulsed APC were required in the IML compartment or inhibition of the PBL response did not occur. Since independence of cell contact suggested that inhibition was mediated by soluble factors, monoclonal antibodies were included in cultures and of these only anti-IL-10 effectively (>90%) reversed the inhibitory effect of IML on PBL responses to purified E. coli proteins identifying this as the cytokine responsible. We found no evidence of IL-2-reversible anergy of IML or of competition at the level of the antigen-presenting cell by culturing cells in different stoichiometries in this system.Conclusion: We have shown in-vitro that human CD4+ IML mediate down-regulation of T-cell responses to commensal bacterial antigens through IL-10 release, although in-vivo other mechanisms such as clonal deletion may also be important.