The lymph nodes draining the small intestine and colon are anatomically separate and immunologically distinct.

Dendritic cells (DCs) in the small intestine (SI) and colon are fundamental to direct intestinal immune responses; they migrate to the mesenteric lymph nodes (MLNs) and prime T cells. We demonstrate anatomical segregation of lymphatic drainage from the intestine, specifically that DCs from the SI and colon migrate to different nodes within the MLN, here called the sMLN and cMLN. As a consequence, different frequencies of DC subsets observed in the SI and colon are reflected among the DCs in the sMLN and cMLN. Consistent with the SI's function in absorbing food, fed antigen is presented in the sMLN, but not in the cMLN. Furthermore, the levels of expression of CCR9 and α4ß7 are increased on T cells in the sMLN compared with the cMLN. DCs from the cMLN and colon are unable to metabolize vitamin A to retinoic acid (RA); thus, DCs may contribute to the differential expression of tissue homing markers observed in the sMLN and cMLN. In summary, the sMLN and cMLN, and the DCs that migrate to these LNs are anatomically and immunologically separate. This segregation allows immune responses in the SI and colon to be controlled independently.

Colonic tolerance develops in the iliac lymph nodes and can be established independent of CD103(+) dendritic cells.

Tolerance to harmless exogenous antigens is the default immune response in the gastrointestinal tract. Although extensive studies have demonstrated the importance of the mesenteric lymph nodes (MLNs) and intestinal CD103(+) dendritic cells (DCs) in driving small intestinal tolerance to protein antigen, the structural and immunological basis of colonic tolerance remain poorly understood. We show here that the caudal and iliac lymph nodes (ILNs) are inductive sites for distal colonic immune responses and that colonic T cell-mediated tolerance induction to protein antigen is initiated in these draining lymph nodes and not in MLNs. In agreement, colonic tolerance induction was not altered by mesenteric lymphadenectomy. Despite tolerance development, CD103(+)CD11b(+) DCs, which are the major migratory DC population in the MLNs, and the tolerance-related retinoic acid-generating enzyme RALDH2 were virtually absent from the ILNs. Administration of ovalbumin (OVA) to the distal colon did increase the number of CD11c(+)MHCII(hi) migratory CD103(-)CD11b(+) and CD103(+)CD11b(-) DCs in the ILNs. Strikingly, colonic tolerance was intact in Batf3-deficient mice specifically lacking CD103(+)CD11b(-) DCs, suggesting that CD103(-) DCs in the ILNs are sufficient to drive tolerance induction after protein antigen encounter in the distal colon. Altogether, we identify different inductive sites for small intestinal and colonic T-cell responses and reveal that distinct cellular mechanisms are operative to maintain tolerance at these sites.

Mechanisms of oral tolerance.

Oral tolerance is the specific immunological unresponsiveness induced by feeding antigen. Although it is an obstacle to oral vaccination, it is probably the mechanism that prevents intestinal hypersensitivity reactions to food antigens and may provide a novel strategy for the treatment of a range of inflammatory disorders. Feeding antigen can provide stable and long-lasting tolerance of a wide range of immune responses to a variety of antigens. However, the mechanisms of oral tolerance and the major factors that influence them remain controversial.

CCR2(+)CD103(-) intestinal dendritic cells develop from DC-committed precursors and induce interleukin-17 production by T cells.

The identification of intestinal macrophages (mφs) and dendritic cells (DCs) is a matter of intense debate. Although CD103(+) mononuclear phagocytes (MPs) appear to be genuine DCs, the nature and origins of CD103(-) MPs remain controversial. We show here that intestinal CD103(-)CD11b(+) MPs can be separated clearly into DCs and mφs based on phenotype, gene profile, and kinetics. CD64(-)CD103(-)CD11b(+) MPs are classical DCs, being derived from Flt3 ligand-dependent, DC-committed precursors, not Ly6C(hi) monocytes. Surprisingly, a significant proportion of these CD103(-)CD11b(+) DCs express CCR2 and there is a selective decrease in CD103(-)CD11b(+) DCs in mice lacking this chemokine receptor. CCR2(+)CD103(-) DCs are present in both the murine and human intestine, drive interleukin (IL)-17a production by T cells in vitro, and show constitutive expression of IL-12/IL-23p40. These data highlight the heterogeneity of intestinal DCs and reveal a bona fide population of CCR2(+) DCs that is involved in priming mucosal T helper type 17 (Th17) responses.

Lymph-borne CD8α+ dendritic cells are uniquely able to cross-prime CD8+ T cells with antigen acquired from intestinal epithelial cells.

Cross-presentation of cellular antigens is crucial for priming CD8(+) T cells, and generating immunity to intracellular pathogens--particularly viruses. It is unclear which intestinal phagocytes perform this function in vivo. To address this, we examined dendritic cells (DCs) from the intestinal lymph of IFABP-tOVA 232-4 mice, which express ovalbumin in small intestinal epithelial cells (IECs). Among lymph DCs (LDCs) only CD103(+) CD11b(-) CD8α(+) DCs cross-present IEC-derived ovalbumin to CD8(+) OT-I T cells. Similarly, in the mesenteric lymph nodes (MLNs), cross-presentation of IEC-ovalbumin was limited to the CD11c(+) MHCII(hi) CD8α(+) migratory DCs, but absent from all other subsets, including the resident CD8α(hi) DCs. Crucially, delivery of purified CD8α(+) LDCs, but not other LDC subsets, into the MLN subcapsular lymphatic sinus induced proliferation of ovalbumin-specific, gut-tropic CD8(+) T cells in vivo. Finally, in 232-4 mice treated with R848, CD8α(+) LDCs were uniquely able to cross-prime interferon γ-producing CD8(+) T cells and drive their migration to the intestine. Our results clearly demonstrate that migrating CD8α(+) intestinal DCs are indispensable for cross-presentation of cellular antigens and, in conditions of inflammation, for the initial differentiation of effector CD8(+) T cells. They may therefore represent an important target for the development of antiviral vaccinations.

Evidence that Ia+ bone-marrow-derived cells are the stimulus for the intestinal phase of the murine graft-versus-host reaction.

We have investigated whether bone marrow (BM) or tissue-derived cells provide the stimulus for the intestinal phase of graft-versus-host reaction (GVHR) in F1 mice injected with parental spleen cells. After injection of CBA cells, the characteristic increases in crypt length, crypt cell production rate (CCPR), and intraepithelial lymphocyte (IEL) count occurred in the jejunum of (CBA x BALB/c)F1----CBA BM chimeras, but not in CBA----(CBA x BALB/c)F1 BM chimeras. Thus, BM-derived cells alone can induce intestinal GVHR and, as similar intestinal alterations were found in (A.TH x A.TL)----A.TL BM chimeric mice after injection of A.TL spleen cells, we deduce that the BM-derived stimulator cells are Ia+. We conclude that the intestinal pathology in this model of GVHR is an indirect effect of soluble mediators that are released during a local delayed-type hypersensitivity (DTH) reaction induced by recognition of Ia+ passenger leukocytes within the mucosa.

Hypersensitivity reactions in the small intestine. 6. Pathogenesis of the graft-versus-host reaction in the small intestinal mucosa of the mouse.

The intestinal phase of the graft-versus-host reaction (GVHR) has been investigated in adult F1 mice bearing grafts of fetal gut using an H-2-incompatible and an H-2-compatible, Mls-incompatible combination. The results indicate that the mitotic activity of the intestinal crypts and the number of intraepithelial lymphocytes are sensitive parameters of the mucosal cell-mediated immune response in the GVHR. Using these indices, indirect evidence has been obtained that soluble factors may be responsible for the damage to the intestine in the GVHR. Possible pathways of lymphocyte activation and migration to the gut are proposed to explain the role of gut-associated T cells in the production of these mediators.

Induction of intestinal graft-versus-host reactions across mutant major histocompatibility antigens by T lymphocyte subsets in mice.

We have examined the ability of highly purified subsets of C57B1/6 L3T4+ and Lyt2+ cells to cause intestinal graft-versus-host reactions in H-2 mutant mice expressing isolated class I (bm1) or II (bm12) MHC mutations. Heavily irradiated B6xbm12)F1 mice given B6 L3T4+ T cells (anti-class II GVHR) developed an acute, lethal GVHR that was associated with intense jejunal crypt hyperplasia and an early rise in the density of intraepithelial lymphocytes. Irradiated (B6xbm1)F1 mice given B6 Lyt2+ T cells (anti-class I GVHR) showed a similar phase of crypt hyperplasia within the first 2 weeks, but this was less marked than for anti-class II GVHR and was not associated with an increase in IEL count. Only very minor gut pathology was observed when B6 Lyt2+ T cells were transferred to irradiated mice carrying the bm9 class I mutation, which is much weaker than the bm1 mutation and does not stimulate Lyt2+ helper T cells. The GVHR mediated by B6 L3T4+ and Lyt2+ T cells was H-2 class-specific, as no pathology was seen in bm12 recipients of Lyt2+ cells or in bm1 recipients of L3T4+ cells. B6 L3T4+ and Lyt2+ T cells both induced splenomegaly and intestinal GVHR in nonirradiated, 4-5-day-old neonatal (B6xbm12)F1 or (B6xbm1)F1 mice, respectively, a form of intestinal GVHR that does not require specific cytotoxic T lymphocytes. Again, Lyt2+ T cells were less efficient than L3T4+ T cells in the induction of GVHR. Thus, both class I and class II MHC-restricted T cells can mediate different forms of intestinal GVHR under appropriate circumstances, but class II MHC-restricted T cells may be more efficient.

Hypersensitivity reactions in the small intestine. VII. Induction of the intestinal phase of murine graft-versus-host-reaction by Lyt 2- T cells activated by I-A alloantigens.

This study has examined the nature of the T lymphocytes and the alloantigens that induce the intestinal phase of graft-versus-host reaction in unirradiated F1 mice. Parental spleen cells were depleted of T cells subsets by treatment with anti-Lyt monoclonal antibodies and complement, and we show that Lyt 2- cells alone induce the increased lymphocytic infiltration of the epithelium that characterizes the intestinal graft-versus-host reaction. Lyt 2- cells are also required to induce some of the associated crypt hyperplasia, but the full crypt changes require both Lyt 2- and Lyt 2+ T cells. In intra-H-2 recombinant congenic F1 mice with graft-versus-host reaction, a disparity at the I-A locus was alone sufficient and necessary for crypt hyperplasia and increased intraepithelial lymphocyte counts, while an I-J incompatibility led to suppression of both these indices. The results support the hypothesis that the intestinal pathology of acute GVHD is induced by class II MHC-restricted delayed-type hypersensitivity effector T cells.

Augmentation of intestinal and peripheral natural killer cell activity during the graft-versus-host reaction in mice.

We have investigated the possibility that nonspecific cytotoxicity may be involved in the pathogenesis of the intestinal phase of the graft-versus-host reaction (GVHR) in mice. A GVHR was induced in unirradiated (CBA X BALB/c)F1 mice and natural killer (NK) cell activity against YAC-1 followed in the spleen, mesenteric lymph node (MLN), and isolated intraepithelial lymphocytes (IEL). Augmented NK activity developed simultaneously in all tissues in parallel with the progress of the GVHR. The NK activity of IEL also showed a close association with the increased numbers of IEL found on sections of small intestine. Mature T lymphocytes and macrophages did not contribute to the nonspecific cytotoxicity, and antihost cytotoxic T cells were not detected in any tissue. The results indicate that generalized recruitment of NK cells occurs during the GVHR both in peripheral and intestinal lymphoid tissues, and we propose that lymphokines are responsible for this phenomenon. NK cells recruited by a delayed-type hypersensitivity reaction may contribute to the pathogenesis of the GVHR, but an alternative explanation is that NK cells may inhibit the progression of the GVHR.

Studies on the immunogenicity of an endogenously processed protein antigen in mice.

We have examined the general immunogenicity of a non-replicating antigen which was introduced artificially into the endogenous pathway of antigen processing. EG7.OVA cells transfected with the OVA gene are efficient presenters of endogenously processed OVA and induced high levels of class I major histocompatibility complex (MHC)-restricted cytotoxic T lymphocytes (CTL) in vivo. In addition, mice immunised with EG7.OVA cells developed immune responses more characteristic of class II MHC-restricted T cells, including IgG antibody production, systemic delayed type hypersensitivity (DTH) and a proliferative response to OVA in vitro. However, most of these responses were small, and EG7.OVA cells did not prime mice for secondary antibody or DTH responses. Thus endogenously synthesised, non-replicating antigens are poor stimulators of T cells which exploit the exogenous processing pathway. If vaccine vectors containing purified epitopes are to stimulate all T cells effectively, they will need to utilise strategies which enable direct entry to both antigen processing pathways.

Oral vaccination with immune stimulating complexes.

There is a need for non-living adjuvant vectors which will induce a full range of local and systemic immune responses to orally administered purified antigens. Here we describe our experience with lipophilic immune stimulating complexes (ISCOMS) containing the saponin adjuvant Quil A. When given orally, ISCOMS containing the model protein antigen ovalbumin (OVA) induce a wide range of systemic immune responses, including Th1 and Th2 CD4 dependent activity, class I MHC restricted cytotoxic T-cell responses and local production of secretory IgA antibodies. More recent results indicate that ISCOMS may act partly by enhancing the uptake of protein from the gut. In addition, intraperitoneal injection of ISCOMS recruits and activates many components of the innate immune system. including neutrophils, macrophages, and dendritic cells. In parallel, there is increased production of nitric oxide (NO), reactive oxygen intermediates (ROI), interleukins (IL) 1, 6, 12, and gamma interferon (gammaIFN). Of these factors, only IL12 is essential for the immunogenicity of ISCOMS in vivo, as mucosal and systemic responses to ISCOMS are reduced in IL12KO mice, but not in IL4KO, IL6KO, inducible NO synthase (iNOS) KO, or gammaIFN receptor KO mice. We propose that ISCOMS act by targetting antigen and adjuvant to macrophages and/or dendritic cells. This pathway may be amenable to exploitation for vaccine development, especially if combined with another vector with a different mucosal adjuvant profile, such as cholera toxin.

Serum paraproteins in chronic lymphocytic leukaemia.

The presence of paraproteins in the sera of 10 patients with chronic lymphocytic leukaemia (CLL) was investigated using immunoisoelectric focusing. Monoclonal immunoglobulins were found in nine of these 10 sera. Five sera contained a single monoclonal IgM paraprotein, one serum contained a single monoclonal IgG paraprotein, while three sera contained more than one monoclonal paraprotein--namely, IgM + IgD, IgM + IgG, and IgM + IgD + IgG. The results indicate that the malignant B cells of CLL may be at a later stage of differentiation than previously assumed and serum monoclonal immunoglobulin could be of value as a tumour marker.

Analysis of the effector functions of different populations of mucosal lymphocytes.

Lymphocytes separated from the epithelial layer of mouse small intestine, IEL, were tested for their NK cytotoxicity against Yac-1 targets. There was little NK activity in a 4 hour assay, but high activity in an 18 hour assay, and the NK activity of IEL did not parallel that in the spleen in any of the mouse strains tested. Furthermore, IEL exerted a suppressor activity on mouse spleen NK activity. Specific T-cell cytotoxicity appeared in IEL in mice immunized with an intraperitoneal injection of P-815 tumor cells. By contrast with IEL, LPL had little NK or NK suppressor activity, but higher levels of specific T-cell cytotoxicity in tumor-immunized mice than intraepithelial lymphocytes. A high proportion of IEL had granules that stained with Giemsa and Astra blue. Furthermore many IEL carried Lyt-2+ phenotype and no other T-cell surface antigen. Intraepithelial lymphocytes appeared, therefore, to have staining and phenotype characteristics of both granular NK cells and suppressor cells. It was clear that the intestinal mucosa contained populations of immune effector cells that were heterogeneous in nature and function.

Inactivation of Th1 and Th2 cells by feeding ovalbumin.

Several different mechanisms have been implicated in oral tolerance to protein antigens, depending on the nature and dose of antigen used and the species under study. Here, we have investigated the basis of unresponsiveness in a well-established model of oral tolerance in mice fed 25 mg ovalbumin (OVA). Our results show that CD8+ T-cell activity is suppressed by feeding OVA and that these cells are not required for the induction of tolerance. CD4+ T cells are essential for tolerance to occur, but both Th1 and Th2 cell-dependent functions are tolerized equally in OVA-fed mice. Peripheral lymph node cells from tolerized mice rapidly undergo apoptosis when cultured in vitro but produce substantial amounts of transforming growth factor beta (TGFbeta) in response to OVA. The appearance of tolerance in vivo is preceded by a transient phase of T-cell priming, and we propose that this model of oral tolerance reflects partial activation of T cells by fed antigen, leading to selective production of TGFbeta and consequent inactivation of all effector T cells. These findings indicate that the active suppression and clonal anergy identified previously in mice with oral tolerance may not be mutually exclusive phenomena.

Induction of proliferative and destructive graft-versus-host reactions in the small intestine.

We have used the intestinal phase of a GvHR to investigate the immunological basis of enteropathies associated with CMI. The nature of the damage depends on the model of GvHR used. Adult unirradiated (CBAxBALB/cF1) mice with GvHR developed an entirely proliferative enteropathy characterized by crypt hyperplasia, increased numbers of IEL and enhanced NK cell activity, but there was no villus atrophy or CTL. Although neonatal or irradiated adult (CBAxBALB/c)F1 mice with GvHR developed a destructive enteropathy with marked villus atrophy and CTL activity, this also required an early proliferative phase identical to that found in unirradiated adult mice. Thus, we propose that proliferative enteropathy in GvHR is due to soluble mediators released by a local DTH reaction but that villus atrophy also requires activation of a further population of effector cells, which may be CTL or suppressor T cells.

The role of antigen processing and suppressor T cells in immune responses to dietary proteins in mice.

We have examined the role of Ts cels and APC in regulating the tolerance of systemic DTH in mice fed OVA. Oral tolerance to OVA was prevented by eliminating Ts with dGuo and by treating mice with anti-I-J antiserum. In addition, activating the reticuloendothelial system (RES) with oestradiol, muramyl dipeptide (MDP) or GvHR prevented the induction of tolerance. Further studies showed that prevention of oral tolerance correlated with the ability to enhance APC activity and that oestradiol also abrogated the induction of Ts after feeding OVA. Our results show that the tolerance of systemic DTH in mice fed OVA reflects complex interactions between APC and Ts and suggest that defects in these regulatory events may be responsible for clinical food sensitive enteropathy.

Dose-range studies of benoxaprofen compared with placebo in patients with active rheumatoid arthritis.

The new nonsteroidal antiinflammatory drug benoxaprofen was compared with placebo in separate double-blind crossover studies in patients with active rheumatoid arthritis given 400 mg, 600 mg, or 800 mg a day. Benoxaprofen was significantly better than placebo at all doses studied. The 800 mg dose was the most effective but there were slightly more gastrointestinal side effects with this dose. The 600 mg dose was more effective than the 400 mg dose with an acceptably low incidence of side effects at both these doses. For most patients, 600 mg once daily will probably be the usual dose of the drug, but some will need the higher dose to 800 mg, while the condition of others may be satisfactorily controlled with just 400 mg daily.

Intestinal CD103(-) dendritic cells migrate in lymph and prime effector T cells.

Intestinal dendritic cells (DCs) continuously migrate through lymphatics to mesenteric lymph nodes where they initiate immunity or tolerance. Recent research has focused on populations of intestinal DCs expressing CD103. Here we demonstrate, for the first time, the presence of two distinct CD103(-) DC subsets in intestinal lymph. Similar to CD103(+) DCs, these intestine-derived CD103(-) DCs are responsive to Flt3 and they efficiently prime and confer a gut-homing phenotype to naive T cells. However, uniquely among intestinal DCs, CD103(-) CD11b(+) CX(3)CR1(int) lymph DCs induce the differentiation of both interferon-γ and interleukin-17-producing effector T cells, even in the absence of overt stimulation. Priming by CD103(-) CD11b(+) DCs represents a novel mechanism for the rapid generation of effector T-cell responses in the gut. Therefore, these cells may prove to be valuable targets for the treatment of intestinal inflammation or in the development of effective oral vaccines.

Oral tolerance to food protein.

Oral tolerance is the state of local and systemic immune unresponsiveness that is induced by oral administration of innocuous antigen such as food proteins. An analogous but more local process also regulates responses to commensal bacteria in the large intestine and, together, mucosally induced tolerance appears to prevent intestinal disorders such as food allergy, celiac disease, and inflammatory bowel diseases. Here we discuss the anatomical basis of antigen uptake and recognition in oral tolerance and highlight possible mechanisms underlying the immunosuppression. We propose a model of stepwise induction of oral tolerance in which specialized populations of mucosal dendritic cells and the unique microenvironment of draining mesenteric lymph nodes combine to generate regulatory T cells that undergo subsequent expansion in the small intestinal lamina propria. The local and systemic effects of these regulatory T cells prevent potentially dangerous hypersensitivity reactions to harmless antigens derived from the intestine and hence are crucial players in immune homeostasis.