Continuing education of the immune system--dendritic cells, immune regulation and tolerance.

T cells, as they develop in the thymus come to express antigen receptors. The specificity of these receptors cannot be predicted and must include many with potential anti-self reactivity. Those that encounter self-antigens, in association with self-MHC (major histocompatibility complex), with high affinity are inactivated and do not leave the thymus. Not all self-antigens however are expressed in the thymus and thus many potentially self-reactive T cells enter the periphery. It poses therefore a fundamental immunological question: how peripheral self-tolerance is maintained in health? Dendritic cells (DC) play a central role in the activation of T cells, especially naïve T cells. Their importance in initiating immune responses against pathogens has been well established. However, DC represent complex populations of cells. Recent advances in our knowledge including molecular understanding of DC/T cell interactions have begun to reveal another important dimension of DC functions in the periphery, being not only initiators but also regulators of the immune system. This review summarises recent findings on the roles of DC in the regulation of immune responses and the maintenance of peripheral tolerance, in an attempt to explain how break down of this may lead to immunopathologies and autoimmunity. The concept of a regulatory DC and its possible role in the generation of T regulatory cells in health and in diseases are also discussed. Based on these, the need for a "continuing education" of the immune system throughout one's life, in which DC are again the "tutors", is postulated.

A discrete subpopulation of dendritic cells transports apoptotic intestinal epithelial cells to T cell areas of mesenteric lymph nodes.

This study identifies a dendritic cell (DC) subset that constitutively transports apoptotic intestinal epithelial cell remnants to T cell areas of mesenteric lymph nodes in vivo. Rat intestinal lymph contains two DC populations. Both populations have typical DC morphology, are major histocompatibility complex class II(hi), and express OX62, CD11c, and B7. CD4(+)/OX41(+) DCs are strong antigen-presenting cells (APCs). CD4(-)/OX41(-) DCs are weak APCs and contain cytoplasmic apoptotic DNA, epithelial cell-restricted cytokeratins, and nonspecific esterase (NSE)(+) inclusions, not seen in OX41(+) DCs. Identical patterns of NSE electrophoretic variants exist in CD4(-)/OX41(-) DCs, intestinal epithelial cells, and mesenteric node DCs but not in other DC populations, macrophages, or tissues. Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick-end labeling (TUNEL)-positive DCs and strongly NSE(+) DCs are present in intestinal lamina propria. Peyer's patches and mesenteric but not other lymph nodes contain many strongly NSE(+) DCs in interfollicular and T cell areas. Similar DCs are seen in the ileum and in T cell areas of mesenteric nodes in gnotobiotic rats. These results show that a distinct DC subset constitutively endocytoses and transports apoptotic cells to T cell areas and suggest a role for these DCs in inducing and maintaining peripheral self-tolerance.

Dendritic cell heterogeneity in vivo: two functionally different dendritic cell populations in rat intestinal lymph can be distinguished by CD4 expression.

DC derived from rat pseudo-afferent lymph (L-DC) vary in CD4, CD11b/c, Thy1, and OX41 expression. CD4 and OX41 are expressed by the same subpopulation (50-60%) of L-DC. CD4+/OX41+ L-DC express short fine processes and low nonspecific esterase, whereas CD4- DC/OX41- express long pseudopodia, high nonspecific esterase, and many cytoplasmic inclusions. These differences are stable in culture. Both populations express similar amounts of MHC class II, ICAM-1, CD11b/c and OX62. Most CD4-/OX41- L-DC are strongly positive for B7, but CD4+ L-DC express less B7, and some may be negative. Both populations express invariant chain, but both the absolute numbers and levels of expression were higher for CD4- DC. Surprisingly, CD4+ L-DC are more potent APC than CD4- cells in MLRs, for sensitized T cells in vitro and for naive T cells in vivo. Cultured CD4+/OX41+ DC can still process and present native Ag. Cultured CD4-/OX41- cells cannot present native Ag but can stimulate strong MLRs. CD4- DC invariant chain expression decreases in culture, whereas expression by CD4+ DC is stable for 48 h. CD4+ and CD4- L-DC have similar turnover times in vivo, suggesting that one population is not the precursor of the other. Thus, two separate DC populations that differ functionally and phenotypically migrate from intestine to mesenteric nodes. This may reflect distinct DC lineages or differentiation modulated by different microenvironmental stimuli.

Dendritic cells interact directly with naive B lymphocytes to transfer antigen and initiate class switching in a primary T-dependent response.

Dendritic cells (DC) are thought to initiate Ab synthesis by activation of T cells, which then provide cytokine and cell-bound "help" to B cells. Here, we provide evidence that DC can capture and retain unprocessed Ag in vitro and in vivo, and can transfer this Ag to naive B cells to initiate a specific Ab response. The response is skewed with 4- to 13-fold higher titers of IgG than IgM, and the predominant subclasses of Ab produced in naive animals are those associated with Th2-type responses. Ag retention and the skew in class switching is a physiologic phenomenon because DC loaded with Ag in vivo and isolated 24 h later initiated a class-switched, Ag-specific Ab response in naive animals. In vitro studies confirmed that DC provide naive B cells with signals that are essential for the synthesis of class-switched Ab. Taken together, these observations show that DC have an important role in the initiation of Ab synthesis by direct interaction with B cells.

Dendritic cells and resting B cells form clusters in vitro and in vivo: T cell independence, partial LFA-1 dependence, and regulation by cross-linking surface molecules.

Initiation of an Ab response requires interaction between dendritic cells (DC), T cells, and B cells in a T cell area. We demonstrate that rat DC and B cells form T cell-independent clusters in vitro and in vivo. In vitro clusters form within 1 h and dissociate within 24 to 48 h. Clustering is restricted to resting B cells, is energy, cytoskeleton, and protein kinase C dependent, and is inhibited by anti-LFA-1 but not anti-ICAM-1 mAbs. Spleen and lymph node B cells cluster more strongly than those from lymph or blood, suggesting up-regulation of adhesiveness during transendothelial migration. Bone marrow B cells do not form clusters. DC from spleen and lymph nodes show the most clustering, lymph-borne DC are intermediate, and DC from lamina propria, Peyer's patches, and those grown from bone marrow form the fewest clusters. Clustering is stimulated by cross-linking MHC class II (whole mAb or F(ab')2) on DC or B cells or Thy-1 on DC, but not MHC class I, CD45, or CD44. Stimulation by mAb is energy, cytoskeletal, and protein kinase C dependent, but is not inhibited by anti-LFA-1 mAbs, suggesting involvement of other, unidentified adhesion molecules. We suggest that interactions between DC and B cells will occur regularly during B cell recirculation. Cross-linking of MHC class II-peptide molecules on DC by specific T cells would increase binding avidity, causing retention of Ag-specific B cells on DC long enough for the B cells to process Ag, thereby facilitating cognate interactions between T and B cells.

Dietary fish oil diminishes the antigen presentation activity of rat dendritic cells.

Rats were fed for 6 weeks on a low fat (LF) diet or on high fat diets containing safflower oil [SO; rich in n-6 polyunsaturated fatty acids (PUFAs)] or fish oil (FO; rich in n-3 PUFAs). Lymph-borne dendritic cells (L-DC) were isolated after cannulation of the thoracic duct and were used as antigen [keyhole limpet hemocyanin (KLH)]-presenting cells in an ex vivo assay that used KLH-sensitized spleen lymphocytes as the responder cells. FO feeding significantly diminished the antigen presentation activity of L-DC compared with L-DC from rats fed each of the other diets. The antigen presentation activity of L-DC from rats fed the SO diet was greater than that of L-DC from rats fed the LF diet. Feeding the FO diet significantly reduced both the proportion of CD2-positive L-DC and the level of CD2 expression on L-DC compared with feeding each of the other diets; the proportions of L-DC staining positive for CD40, CD18, CD54, CD11a, and MHC II were unaffected by diet. However, FO feeding reduced the level of expression of CD18, CD11a, MHC II, and CD54 on L-DC compared with feeding the other two diets; the level of expression of CD40 was unaffected by diet. This is the first study to report effects of dietary fatty acids on dendritic cells. The suppressive effect of FO feeding may account for some of the beneficial effects of n-3 polyunsaturated fatty acids observed in clinical settings, such as prolonged survival of grafts and diminished chronic inflammatory responses. However, such an effect may also be detrimental because host defense toward bacterial and other antigens could be compromised.

Rat intestinal dendritic cells: immunostimulatory potency and phenotypic characterization.

Dendritic cells (DC) acquire antigens in peripheral tissues, transport them to lymph nodes and present peptides to T cells. DC are particularly good activators of resting T cells. Murine Langerhans' cells (LC) are efficient at endocytosing and processing antigens but are very weak immunostimulators. In culture LC lose the ability to process antigen and become potent immunostimulators. Other peripheral DC are not well characterized and it is not known if they are similarly weak immunostimulators. We isolated DC from rat Peyer's patches (PP) and lamina propria (LP) of the small intestine, from intestinal lymph (LDC) and mesenteric lymph nodes, and examined their ability to stimulate an allogeneic mixed leucocyte reaction (MLR). Freshly isolated LP DC and PP DC could stimulate a moderate MLR but fresh LDC were significantly more potent. After overnight culture, LDC did not change their potency but DC from LP and PP became as potent as LDC. In contrast, fresh lymph node DC stimulated a MLR or oxidative mitogenesis as efficiently as LDC. These results show that the weak immunostimulation of murine LC is not characteristic of all peripheral DC. We compared the phenotypes of DC from different sites before and after culture. Different populations of DC show marked phenotypic heterogeneity in the expression of surface markers, particularly Thy-1, CD2 and the iC3b receptor. PP and LP DC were similar to MLN DC in their expression of markers, but differed from LDC. After culture there were marked changes in DC surface marker expression and the differences between the populations were reduced. These observations suggest that the heterogeneity observed in fresh populations does not signify different stages of maturation but may represent activation.

Antigen processing: cultured lymph-borne dendritic cells can process and present native protein antigens.

Langerhans' cells (LC) cultured for 1-3 days lose their ability to process native protein antigens but acquire the ability to stimulate resting T cells as assessed in an allogeneic mixed lymphocyte response (MLR). Lymph-borne dendritic cells (L-DC) are physiologically involved in the transport of antigens to lymph nodes but it is not known whether these cells lose the ability to process antigens in culture. To investigate this, we cultured L-DC derived from the intestine for 20-72 hr and tested their ability to process and present antigens. Our results show that these L-DC are able to present antigen to primed spleen T cells as effectively as fresh cells. To exclude the possibility that commercial ovalbumin (OVA) preparations contain peptides which might bind directly to major histocompatibility complex (MHC) molecules, OVA was filtered through Sephadex G50 and the peak fractions used as antigen. The results show that cultured L-DC are also able to present G50-filtered OVA efficiently to primed spleen T cells. More importantly, these G50-OVA-pulsed L-DC are able to prime naive T cells specifically in vivo. Chloroquine inhibited the ability of both fresh and cultured L-DC to present antigen to primed T cells but did not inhibit their ability to stimulate a MLR, indicating that processing was a necessary step for antigen presentation. Taken together, these results clearly show that cultured L-DC are active in processing and presenting native antigens and the hypothesis proposed for LC does not apply to rat lymph-borne dendritic cells. The physiological significance of these observations is discussed.

Endotoxin-mediated dendritic cell release from the intestine. Characterization of released dendritic cells and TNF dependence.

Dendritic cells (DC) acquire Ag in peripheral tissues and transport it to lymph nodes where they efficiently activate resting T cells. We have shown that i.v. endotoxin causes increased release of intestinal DC into lymph. In this paper we further characterize the release of DC and the properties of the released cells. A total of 50 micrograms of endotoxin injected i.v. causes an increase in DC output within 6 h that peaks between 12 and 24 h, with a maximum output of 8 to 15 times normal. At the same time lymphocyte output is markedly decreased. The increased output of DC is followed by a decrease to subnormal levels. The stimulated release of DC is almost totally blocked by a monoclonal anti-TNF-alpha Ab. A second injection of TNF-alpha does not result in further DC release. DC are not released from lymph nodes into efferent lymph by endotoxin. DC collected from lymph after endotoxin treatment show increased expression of the p55 IL-2 receptor and the OX48 Ag but otherwise resemble normal lymph DC. In functional assays they show no significant differences from normal in their ability to stimulate a MLR or to present Ags to sensitized T cells. Immunocytochemistry with the use of MRC OX62 suggests that the DC are released into lymph from the lamina propria of the small intestine. The stimulated release of DC mediated by TNF-alpha may be important in regulating Ag presentation in lymph nodes draining inflammatory sites.

Rat macrophage lysosomal membrane antigen recognized by monoclonal antibody ED1.

The monoclonal antibody (mAb) ED1 is being used widely as a marker for rat macrophages. The distribution of the recognized antigen in tissues and isolated cells strongly supports this use as a macrophage marker, since the majority of macrophages are recognized and only seldomly are other cell types stained by mAb ED1. In the present study we further characterized the recognized antigen by a detailed description of the localization of the antigen and by determining biochemical and functional properties. We show that the antigen is expressed on the membranes of cytoplasmic granules, like phagolysosomes, as well as on the cell surface. The amount of ED1 expression in a single cell can be correlated to phagocytic activity of the respective cell type, but the mAb ED1 is not able to block latex phagocytosis or bacterial killing. The mAb ED1 appears to recognize a heavily glycosylated protein of 90,000-110,000 MW, depending on the cell type used as antigen source. A possible relation with other known lysosomal glycoproteins with a similar molecular weight is discussed.

Antigen acquisition by dendritic cells: intestinal dendritic cells acquire antigen administered orally and can prime naive T cells in vivo.

In the rat, mesenteric lymphadenectomy allows collection of dendritic cells (DC) derived from the small intestine after cannulation of the thoracic duct. We prepared rats this way and administered antigens by oral feeding or intraintestinal injection. DC enriched from the thoracic duct lymph collected over the first 24 h from these animals are able to stimulate sensitized T cells in vitro and to prime popliteal lymph node CD4+ T cells after footpad injection, while B and T cells from the same thoracic duct lymph are inert in priming. 500 or less DC pulsed in vitro with antigen can prime T cells in vivo, whereas 100 times more B cells or macrophages pulsed in vitro are quite inert. 1 mg of ovalbumin administered orally is sufficient to load DC for in vivo priming of T cells. Antigen could not be detected directly in DC but was present in macrophages in the lamina propria. Direct presentation of antigen by DC to T cells was demonstrated by injecting F1 recipients with parental DC and showing restriction of T cell sensitization to the major histocompatibility complex of the injected DC. Antigen-bearing DC do not induce a detectable primary antibody response but a small secondary antibody response can be detected after a boosting injection. These results show that acquisition of antigens by DC in the intestine is very similar to what occurs in vitro or in other tissues, suggesting that there may be no special difference in antigen handling at mucosal surfaces. One implication of these results is that hypotheses designed to explain oral tolerance must take into account the presence of immunostimulatory, antigen-bearing DC in animals that have received oral antigens.

Lymph-borne (veiled) dendritic cells can acquire and present intestinally administered antigens.

To investigate the ability of lymph-borne (veiled) dendritic cells (LDC) to acquire and present antigens in vivo, mesenteric lymphadenectomized rats were injected intra-intestinally with antigen and LDC were then purified from thoracic duct lymph. When used as antigen presenting cells with primed spleen cells as responders, the LDC could stimulate antigen-specific proliferation of the responder cells in the absence of exogenous added antigen. As little as 10 mg of ovalbumin (OVA) or horseradish peroxidase (HRP) injected into the ileum and jejunum could sensitize LDC for presentation. LDC acquired antigen within 8 hr of its injection but cells collected more than 24 hr after injection were unable to stimulate a response. Non-dendritic cells (NDC) in the thoracic duct lymph, such as B cells, were unable to present antigen either following intraintestinal injection or after in vitro pulsing. The antigen-specific response was blocked by antibodies to CD4 and major histocompatibility complex (MHC) class II and was totally dependent on the presence of CD4+ cells in the responding population. These studies show that dendritic cells can acquire antigens in vivo and provide a novel approach to the study of intestinal immune responses and oral tolerance.

Analysis of thymus-independent type 2 antigen uptake by marginal zone macrophages in thin slices of viable lymphoid tissue in vitro.

There is evidence to suggest that the B cell population in the marginal zone (MZ) of the spleen is responsible for the antibody response to thymus-independent type 2 antigens (TI-2). The macrophage (M phi) population in the MZ has been shown to take up TI-2 antigens selectively, and this uptake is potentially important in understanding antigen handling in TI-2 responses. Uptake has not been studied in vitro because isolation of MZ M phi completely abrogates uptake. We have adapted a technique for cutting thin slice of viable spleen which retained this M phi function under in vitro conditions and allowed manipulation of the system. This technique may have widespread application in the study of tissue M phi which are difficult to isolate. Using this method, we show that MZ M phi in splenic slices in vitro selectively take up a TI-2 antigen, in a way apparently identical to that seen in vivo. This is not due to high pinocytic activity by these cells nor to their anatomical location. We provide evidence for a receptor-mediated uptake system, whose function is sensitive to collagenase. The ligand specificity of the uptake system showed unexpected cross-reactivity with the mannosyl-fucosyl receptor with high affinity for mannan, but was not identical to it, and may be indicative of the natural ligands for the TI-2 antigen uptake system.