Altered dendritic cells (DC) might be responsible for regulatory T cell imbalance and autoimmunity in nonobese diabetic (NOD) mice.

Nonobese diabetic (NOD) mice spontaneously develop diabetes, an auto-immune disease characterized by the destruction of insulin-secreting beta-cells by autoreactive T cells. Defects in development and/or functions of dendritic cells (DC) might be critical in eliciting the auto-immune reaction to beta cells in this model. In this paper, DC differentiation in NOD mice was investigated in vitro using bone marrow-derived progenitors (BM-DC) in the presence of GM-CSF and IL-4 or spleen-derived progenitors in the presence of GM-CSF and early acting cytokines such as Flt-3L and IL-6 (SPL-DC). In both culture systems, the absolute number of NOD DC generated was strongly reduced as compared to control strains. In addition, both BM-DC and SPL-DC from NOD mice show defective differentiation into mature DC in conventional culture conditions as indicated by low expression of MHC class II and CD80 molecules among CD11c positive cells and low capacity to stimulate allogeneic T cells. However, DC achieved full maturation when exposed to LPS, except for MHC class II expression that remained decreased. Ex vivo analysis confirmed an unusual phenotype of NOD DC. Both sets of results are thus consistent with a specific defect of DC maturation in these mice.

T lymphocytes potentiate murine dendritic cells to produce IL-12.

IL-12 is mainly produced by CD8alpha(+) dendritic cells (DCs) and induces Th1 polarization of the immune response. We investigated the influence of lymphocytes on splenic DC (SDC) and thymic DC (TDC) development and on their IL-12 production capacity. First, CD3epsilon(-/-) mice, lacking T cells, and RAG-2(-/-) mice, lacking T and B cells, possess numbers of SDCs, TDCs, and CD8alpha(+) SDCs similar to wild-type (WT) mice. Second, SDCs and TDCs from CD3epsilon(-/-) mice do not secrete IL-12 in vitro after different stimulations, whereas DCs from pTalpha(-/-) mice, possessing reduced T cell number, and RAG-2(-/-) mice, produce an IL-12 level similar to that of WT DCs. We show that T lymphocytes restore the capacity of DCs to produce IL-12 after stimulation in vivo by reconstitution of CD3epsilon(-/-) mice with WT T cells and in vitro by coculture of CD3epsilon(-/-) DCs with WT T cells. The regulation of IL-12 production occurred at the transcriptional level, with an increase of IL-12p35 transcripts and a decrease of IL-12p40 transcripts. Although IL-4 restores IL-12 production by CD3epsilon(-/-) SDCs, anti-IL-4 Abs inhibited only partially the IL-12 production in coculture of CD3epsilon(-/-) DCs and WT T cells. Taken together, these data show that T lymphocytes potentiate IL-12 production by DCs and that IL-4 is not solely involved in this regulation. In conclusion, B and T cells exert balanced actions on DCs by respectively inhibiting or promoting IL-12 production.

Granulocyte-dendritic cell unbalance in the non-obese diabetic mice.

Several investigators, including ourselves, have reported lower yield of GM-CSF bone marrow-derived dendritic cells (DC) with altered MHC class II and co-stimulatory molecules expression in the non-obese diabetic (NOD) mice. However, whether this defect was intrinsic to the DC lineage and/or related to abnormal expansion of other cell types responding to GM-CSF remained an opened issue. We performed phenotypical and morphological analysis of cells from GM-CSF-supplemented-bone marrow-cultures and of freshly isolated bone marrow and blood cells from unmanipulated prediabetic NOD mice. The results show a heretofore undescribed bias towards generation of granulocytes in NOD mice, concomitant with quantitative and qualitative alterations of the DC lineage in both the bone marrow and the blood of this mouse strain. We propose that increased generation of granulocytes in NOD mice might contribute to autoimmunity. First, high numbers of granulocytes per se might favor inflammatory environment. Second, granulocytes, by interfering with DC development, might favor unbalanced antigen presenting cell function leading to T cell autoimmunity.

The cytoplasmic domain of Mpl receptor transduces exclusive signals in embryonic and fetal hematopoietic cells.

The Mpl receptor plays an important role at the level of adult hematopoietic stem cells, but little is known of its function in embryonic and fetal hematopoiesis. We investigated the signals sent by the MPL cytoplasmic domain in fetal liver hematopoietic progenitors and during embryonic stem (ES) cell hematopoietic commitment. Mpl was found to be expressed only from day 6 of ES cell differentiation into embryoid bodies. Therefore, we expressed Mpl in undifferentiated ES cells or in fetal progenitors and studied the effects on hematopoietic differentiation. To avoid the inadvertent effect of thrombopoietin, we used a chimeric receptor, PM-R, composed of the extracellular domain of the prolactin receptor (PRL-R) and the transmembrane and cytoplasmic domains of Mpl. This allowed activation of the receptor with a hormone that is not involved in hematopoietic differentiation and assessment of the specificity of responses to Mpl by comparing PM-R with another PRL-R chimeric receptor that includes the cytoplasmic domain of the erythropoietin receptor (EPO-R) ([PE-R]). We have shown that the cytoplasmic domain of the Mpl receptor transduces exclusive signals in fetal liver hematopoietic progenitors as compared with that of EPO-R and that it promotes hematopoietic commitment of ES cells. Our findings demonstrate for the first time the specific role of Mpl in early embryonic or fetal hematopoietic progenitors and stem cells.

Fibroblasts inhibit the production of interleukin-12p70 by murine dendritic cells.

Interleukin-12 p70 (IL-12p70) is a key cytokine produced by dendritic cells (DC) able to drive the development of T helper type 1 (Th1) lymphocytes. We showed that thymic and other fibroblasts strongly inhibit IL-12p70 production by splenic DC stimulated by lipopolysaccharide plus either anti-CD40 or interferon-gamma (IFN-gamma) and by purified splenic DC stimulated by Pansorbin plus IFN-gamma. This IL-12p70 inhibitory activity is secreted in the conditioned medium of primary fibroblasts and fibroblast cell lines but not by haematopoietic cell lines. As IL-10 was the unique factor able to inhibit IL-12p70 produced by cultured splenic DC, we showed that a neutralizing antibody to IL-10 did not suppress the IL-12p70 inhibitory activity of thymic fibroblast-conditioned medium (FCM). This FCM potently inhibits the maturation and expression of major histocompatibility complex class II and co-stimulatory molecules induced by stimulation of spleen-derived DC. While thymic FCM suppressed the IL-12p70 expression by stimulated spleen-derived DC, tumour necrosis factor-alpha production is not affected. This inhibitory activity is able to down-regulate the IL-12p35 subunit transcription and expression, resulting in the impaired assembly of IL-12p70 heterodimer. As fibroblasts are present in the tissue microenvironment and are active players in the establishment of an immune response, the nature and role of the fibroblastic inhibitory activity remain to be established.

Cryptic O2- -generating NADPH oxidase in dendritic cells.

All the components of the O(2)(-)-generating NADPH oxidase typically found in neutrophils, namely a membrane-bound low potential flavocytochrome b and oxidase activation factors of cytosolic origin, are immunodetectable in murine dendritic cells (DCs). However, in contrast to neutrophils, DCs challenged with phorbol myristate acetate (PMA) can barely mount a significant respiratory burst. Nevertheless, DCs generate a substantial amount of O(2)(-) in the presence of PMA following preincubation with pro-inflammatory ligands such as lipopolysaccharide and pansorbin, and to a lesser extent with anti-CD40 or polyinosinic polycytidylic acid. We found that the virtual lack of the oxidase response to PMA alone is specifically controlled in DCs. Through the use of homologous and heterologous cell-free systems of oxidase activation, we showed the following: (1) a NADPH oxidase inhibitory factor is located in DC membranes; it exerts its effect on oxidase activation and not on the activated oxidase. (2) The inhibition is relieved by pretreatment of DC membranes with beta-octylglucoside (beta-OG). (3) The beta-OG-extracted inhibitory factor prevents the activation of neutrophil oxidase. (4) The inhibitory activity is lost after treatment of DC membranes with proteinase K or heating, which points to the protein nature of the inhibitory factor. Overall, these data indicate that the O(2)(-)-generating oxidase in DCs is cryptic, owing to the presence of a membrane-bound inhibitor of protein nature that prevents oxidase activation. The inhibition is relieved under specific conditions, including a prolonged contact of DCs with pro-inflammatory ligands from microbial origin, allowing a substantial production of O(2)(-), which may contribute to the response of DCs to a microbial exposure.