[Human thymic dendritic cells]. / Les cellules dendritiques thymiques humaines.

Human thymic dendritic cells (DC) were enriched from thymocyte suspension by low density fractionation and elimination of CD3, CD8, CD19, CD56 and CD34-positive cells. Flow cytometry analysis shows that they belong to two distinct populations. The prominent (2/3) one is CD4+ CD45RA+ CD33-CD11c-, mostly immature as it lacks CD80, CD83 and CD86, and is HLA-DRint. High expression of CD123 and expression of pre-TCR-alpha transcripts links this subset to lymphoid, interferon-producing cells. The other DC are typically CD4+ CD33+ CD11c+ myeloid cells mostly mature CD83+ HLA-DRhi. Both subsets could be generated in vitro from thymic CD34+ progenitors. Finally although experimental evidence ascribed to thymic DC a major role in establishing central tolerance through deletion of self reactive thymocytes, the respective functions of human lymphoid and myeloid subsets in the human thymus remain to be established.

Identification of mature and immature human thymic dendritic cells that differentially express HLA-DR and interleukin-3 receptor in vivo.

We have previously shown that thymic CD34+ cells have a very limited myeloid differentiation capacity and differentiate in vitro mostly into CD1a+-derived but not CD14+-derived dendritic cells (DC). Herein we characterized the human neonatal thymic DC extracted from the organ in relationship with the DC generated from CD34+ cells in situ. We show that in vivo thymic DC express E cadherin, CLA, CD4, CD38, CD40, CD44, and granulocyte-macrophage colony-stimulating factor-R (GM-CSF-R; CD116) but no CD1a. According to their morphology, functions, and surface staining they could be separated into two distinct subpopulations: mature HLA-DRhi, mostly interleukin-3-R (CD123)-negative cells, associated with thymocytes, some apoptotic, and expressed myeloid and activation markers but no lymphoid markers. In contrast, immature HLA-DR+ CD123hi CD36+ cells with monocytoid morphology lacked activation and myeloid antigens but expressed lymphoid antigens. The latter express pTalpha mRNA, which is also found in CD34+ thymocytes and in blood CD123hi DC further linking this subset to lymphoid DC. However, the DC generated from CD34+ thymic progenitors under standard conditions were pTalpha-negative. Thymic lymphoid DC showed similar phenotype and cytokine production profile as blood/tonsillar lymphoid DC but responded to GM-CSF, and at variance with them produced no or little type I interferon upon infection with viruses and did not induce a strict polarization of naive T cells into TH2 cells. Their function in the thymus remains therefore to be elucidated.

Functional and phenotypic analysis of thymic CD34+CD1a- progenitor-derived dendritic cells: predominance of CD1a+ differentiation pathway.

Whether thymic dendritic cells (DC) are phenotypically and functionally distinct from the monocyte lineage DC is an important question. Human thymic progenitors differentiate into T, NK, and DC. The latter induce clonal deletion of autoreactive thymocytes and therefore might be different from their monocyte-derived counterparts. The cytokines needed for the differentiation of DC from thymic progenitors were also questioned, particularly the need for GM-CSF. We show that various cytokine combinations with or without GM-CSF generated DC from CD34+CD1a- but not from CD34+CD1a+ thymocytes. CD34+ thymic cells generated far fewer DC than their counterparts from the cord blood. The requirement for IL-7 was strict whereas GM-CSF was dispensable but nonetheless improved the yield of DC. CD14+ monocytic intermediates were not detected in these cultures unless macrophage-CSF (M-CSF) was added. Cultures in M-CSF generated CD14-CD1a+ DC precursors but also CD14+CD1a- cells. When sorted and recultured in GM-CSF, CD14+ cells down-regulated CD14 and up-regulated CD1a. TNF-alpha accelerated the differentiation of progenitors into DC and augmented MHC class II transport to the membrane, resulting in improved capacity to induce MLR. The trafficking of MHC class II molecules was studied by metabolic labeling and immunoprecipitation. MHC class II molecules were transported to the membrane in association with invariant chain isoforms in CD14+ (monocyte)-derived and in CD1a+ thymic-derived DC but not in monocytes. Thus, thymic progenitors can differentiate into DC along a preferential CD1a+ pathway but have conserved a CD14+ maturation capacity under M-CSF. Finally, CD1a+-derived thymic DC and monocyte-derived DC share very close Ag-processing machinery.

Theophylline-induced B-CLL apoptosis is partly dependent on cyclic AMP production but independent of CD38 expression and endogenous IL-10 production.

Our recent work has shown that theophylline which inhibits intracellular cyclic adenosine monophosphate (cAMP) degradation is able to kill chronic lymphocytic leukemia (CLL) cells in vitro and synergizes in vitro and in vivo with chlorambucil. In order to test the hypothesis that theophylline works through an indirect increase in cAMP, we have investigated the role of several molecules on B-CLL cells from 20 patients. Direct cAMP inducers such as dibutyryl-cAMP (db-cAMP), prostaglandin-E2 (PGE2) and forskolin induced moderate apoptosis but extremely high levels of intracellular cAMP. By contrast theophylline was highly apoptotic but did not synergize with cAMP inducers. Apoptosis was completely reversed by a cAMP antagonist when induced by PGE2 or forskolin, but was only partially antagonized when induced by theophylline. Since CD38+ CLL cells are more sensitive to apoptosis and since CD38 is enhanced by cAMP inducing agents its expression was investigated. In our hands CD38 was not induced by the above pharmacological compounds. Exogenous IL-10 has been shown to induce CLL cell death; however, apoptosis following treatment with theophylline or cAMP inducers could not be ascribed to endogenous production of IL-10. This ruled out the involvement of cytokines or of an activation or differentiation process in apoptosis. Altogether our data show that an increase in intracellular cAMP mediates apoptosis in vitro but accounts only partly for theophylline-mediated apoptosis.

Human CD34(+) thymocyte maturation: pre-T and NK cell differentiation on neonatal thymic stromal cell culture.

The development of a simple assay for studying human T and NK lymphopoiesis from CD34 progenitors is of great interest. T and NK cells arise from a common CD34(+) immature precursor. While T cell maturation is dependent on interactions and cytokines provided by thymic stromal cells. NK maturation does not require the thymic microenvironment and primarily takes place extrathymically. In addition to models using a mouse thymic microenvironment, in vitro assays based on coculture on human fetal thymic stroma have been described. As an alternative source of fetal thymic tissue we studied the capacity of neonatal thymic epithelial cell enriched stroma to support T and NK cell differentiation. While in the fetal-based assays on NK cells were observed under the conditions used for T cell differentiation, neonatal stroma can generate CD3/TcR+ as well as CD56(+)3(-)8(+)NKR-P1(+)NK cells from both CD34(+)1(-) and CD34(+)1(+) thymocyte precursors. However, following acquisition of CD3/TcR, T-lineage cells disappeared from the culture after 2-3 weeks as a consequence of the outgrowth of the NK cells. These CD56(+)3(-) NK cells appeared to be functionally immature as they required incubation with IL2 or IL15 to lyse K562 target cells. Our data offer a simple and reliable assay for studying the reconstitution potential of T and NK cell progenitors on a monolayer of thymic epithelial cells.

Adoptively transferred CD4+ lymphocytes from CD8 -/- mice are sufficient to mediate the rejection of MHC class II or class I disparate skin grafts.

Recent studies revealed that CD4+ cells initiate allograft rejection through direct recognition of allogeneic MHC class II Ags and indirect recognition of MHC peptides processed by self APCs. Both pathways were shown to help CD8+ cells that eventually lysed allogeneic MHC class I-presenting targets. There was little evidence, however, that CD4+ cells are sufficient for graft rejection. We studied skin graft rejection by CD8-deficient (CD8 -/-) mice. We showed that BALB/cJ(H-2d) CD8 -/- mice could reject allogeneic skin from C57BL/6J(H-2b) mice deficient in MHC class I or in MHC class II Ags. To understand the role of CD4+ cells in this process, we isolated them from CD8 -/- mice and transferred them to BALB/cJ nude mice that had been grafted with allogeneic skin (H-2b) from animals deficient in MHC class I or MHC class II. Nude mice injected with CD4+ cells rejected MHC class II and, albeit more slowly, MHC class I disparate skins. We showed in vitro evidence that CD4+ cells were not cytotoxic toward MHC class I or MHC class II disparate targets and that they recognized MHC class I allogeneic targets through indirect recognition. CD4+ cells produced Th1 cytokines, but not IL-4, following stimulation with allogeneic cells. Furthermore, intragraft TNF-alpha was elevated in skin grafted onto nude mice reconstituted with CD4+ cells compared with nonreconstituted mice. This suggests that MHC class II- or MHC class I-guided CD4+ cells alone are sufficient to induce rejection by the generation of cytokine-induced lesions.

CD4-negative cytotoxic T cells with a T cell receptor alpha/beta intermediate expression in CD8-deficient mice.

Targeted disruption of the CD8 gene results in a profound block in cytotoxic T cell (CTL) development. Since CTL are major histocompatibility complex (MHC) class I restricted, we addressed the question of whether CD8-/- mice can reject MHC class I-disparate allografts. Studies have previously shown that skin allografts are rejected exclusively by T cells. We therefore used the skin allograft model to answer our question and grafted CD8-/- mice with skins from allogeneic mice deficient in MHC class II or in MHC class I (MHC-I or MHC-II-disparate, respectively). CD8-/- mice rejected MHC-I-disparate skin rapidly even if they were depleted of CD4+ cells in vivo (and were thus deficient in CD4+ and CD8+ T cells). By contrast, CD8+/+ controls depleted of CD4+ and CD8+ T cells in vivo accepted the MHC-I-disparate skin. Following MHC-I, but not MHC-II stimulation, allograft-specific cytotoxic activity was detected in CD8-/- mice even after CD4 depletion. A population expanded in both the lymph nodes and the thymus of grafted CD8-/- animals which displayed a CD4-8-3intermediateTCR alpha/betaintermediate phenotype. Indeed its T cell receptor (TCR) density was lower than that of CD4+ cells in CD8-/- mice or of CD8+ cells in CD8+/+ mice. Our data suggest that this CD4-8- T cell population is responsible for the CTL function we have observed. Therefore, MHC class I-restricted CTL can be generated in CD8-/- mice following priming with MHC class I antigens in vivo. The data also suggest that CD8 is needed to up-regulate TCR density during thymic maturation. Thus, although CD8 plays a major role in the generation of CTL, it is not absolutely required.

Early human thymocyte proliferation is regulated by an externally controlled autocrine transforming growth factor-beta 1 mechanism.

Early thymocytes undergo extensive proliferation after their entry into the thymus, but cellular interactions and cytokines regulating this intrathymic step remain to be determined. We analyzed the effects of various T-cell growth factors and cellular interactions on in vitro proliferation of early CD2+CD3/TCR-CD4-CD8- (triple negative [TN]) human thymocytes. Freshly isolated TN cells were then assayed for their growth capacity after incubation with CD2I+III-monoclonal antibody (MoAb), recombinant human interleukin-2 (IL-2), IL-7, and/or IL-4. These cells displayed significant proliferative responses with IL-4, IL-7, or CD2-MoAb+IL-2. The addition of recombinant transforming growth factor beta (TGF beta) or autologous irradiated CD3+CD8+CD4- cells to TN cell cultures dramatically decreased their growth responses to IL-2 and IL-7, whereas IL-4-induced proliferation was less sensitive to growth inhibition. We thus asked whether the CD8+ cell-derived inhibitory effect was due to TGF beta. The addition of neutralizing anti-TGF beta MoAb completely abolished CD8+ cell-derived inhibition of TN cell growth. Analysis of CD8+ cell-derived supernatants indicated that these cells had low TGF beta 1 production capacity, whereas TN cells secrete significantly high levels of TGF beta 1. Cell fixation studies showed that TN cells were the source of the TGF beta. TGF beta 1 released from TN cells was in the latent form that became the active inhibitory form through interaction of TN cells with CD8+ cells. Together, these data suggest a role for TGF beta 1 as an externally controlled, autocrine inhibitory factor for human early thymocytes, with a regulatory role in thymic T-cell output.

[Phenotypic and functional characterization of human prothymocytes]. / Caractérisation phénotypique et fonctionnelle des prothymocytes humains.

We report herein the phenotypic and functional analysis of human bone marrow and thymus derived early T cells. Commitment to T cell lineage is acquired during CD7 antigen expression by CD34+ precursors in human bone marrow and before thymus colonization. Early thymocytes show similar phenotypic characteristics as bone marrow T cells. They rapidly acquire CD4 before the dual expression of CD4 and CD8. Their expansion and differentiation is regulated by two major factors: thymic stroma and cytokines produced by these stroma cells or by thymocytes themselves. Among cytokines, IL1 and sCD23 produced by thymic epithelial cells support in vitro early T cell development.

Cytokine effects of CD23 are mediated by an epitope distinct from the IgE binding site.

Human CD23 and its soluble forms (sCD23) display various biological activities, in addition to their IgE binding function (IgE/BF). The IgE binding domain was recently mapped to residues between Cys163 and Cys282 but its involvement in IgE-independent, CD23 functions remains unknown. In order to clarify this point, a series of N-terminal, C-terminal and internal deletion mutants of CD23 or sCD23 were expressed in CHO cells and tested for their ability (i) to bind to IgE, (ii) to induce colony formation by human myeloid precursor cells, (iii) to promote mature T cell marker expression by early prothymocytes, and (iv) to regulate IgE synthesis. The present study indicates that cytokine activities require the presence of Cys288, while this amino acid is not necessary for IgE/BF. Blocking experiments using various conformation-sensitive monoclonal antibodies further suggest that active epitope(s) of CD23 in cytokine assays is(are) distinct from those involved in IgE/BF.

Epidermal keratinocyte-derived basophil promoting activity. Role of interleukin 3 and soluble CD23.

Human epidermal keratinocytes (EK) secrete factors able to sustain the proliferation of early myeloid cells and, in particular, the generation of basophils. This activity was previously attributed to IL-3, although no definitive in situ demonstration of this cytokine was provided. In regard to the possible physiological relevance of these data, we investigated herein the nature of EK-derived factors responsible for basophil promotion. Our data show that EK-derived supernatants (EK-sup) contain IL-3 as well as soluble CD23 (sCD23), both known for their colony stimulating activity. Messenger RNA for IL-3 and CD23 were also detected in EK. Blocking experiments using specific neutralizing monoclonal antibodies (mAb) further indicate that EK-derived basophil promoting activity is mainly due to the presence of IL-3 and sCD23 in EK-sup. Furthermore, by contrast to IL-3, sCD23 secretion by EK is cortisone sensitive and highly enhanced by IL-4, suggesting distinct regulatory mechanisms for their production.

Thymic epithelial cell-derived supernatants sustain the maturation of human prothymocytes: involvement of interleukin 1 and CD23.

During their development, human CD7+ lymphoid stem cells migrate into the thymus where, following intimate contact with thymic tissue, they proliferate and differentiate into functionally mature T lymphocytes. In this study, we investigated the effect of thymic epithelial cell-derived supernatants (TEC-SN) on early CD7+CD2-CD3- thymocytes. Our results indicate that TEC-SN are able to promote CD2 and CD3/TcR alpha/beta expression by CD7+ precursors. This activity correlated with soluble CD23 and interleukin 1 levels in TEC-SN. Furthermore, monoclonal antibodies to these cytokines decreased in vitro maturation of prothymocytes. Thus, in addition to cell-cell interactions, human TEC produce cytokines able to support early steps of thymocyte differentiation.

Synergistic effect of interleukin 1 and soluble CD23 on the growth of human CD4+ bone marrow-derived T cells.

Low-affinity Fc epsilon receptor (Fc epsilon RII/CD23) is expressed by various human cells and known to be cleaved into soluble fragments (sCD23). Several biological activities were ascribed to these molecules. In this study, we have assessed the effect of recombinant 25-kDa sCD23 (rsCD23) on human bone marrow-derived T cells. Our results show that rsCD23 in synergy with recombinant interleukin 1 enhances mitogenic responsiveness of CD4+ T cells but does not affect CD8+ cell growth. Furthermore, rsCD23 synergizes autologous marrow cells in enhancement of CD4+ cell growth while CD23 monoclonal antibodies decrease accessory cell effect. Together, these data confirm cytokine-like activity of sCD23 on human T cell lineage.

[Soluble CD23: cytokine actives on human hematopoietic precursors]. / Le CD23 soluble: une cytokine active sur les précurseurs hématopoïétiques humains.

We report herein the effect of soluble CD23 (sCD23) on the differentiation of early lymphoid and myeloid human precursors. CD23 is known as a low affinity receptor for IgE. In addition, our results show that sCD23 in synergy with IL1 has a potent activity on the maturation of prothymocytes and the proliferation of multipotential normal and leukemic myeloid precursors.

Soluble CD23 increases IL-3 induction of histamine synthesis by human bone marrow cells.

Interleukin 3 (IL-3) and soluble CD23 (sCD23/IgERII) have similar capacities to induce basophil development from human bone marrow (BM) precursors. IL-3 also has the capacity to induce de novo histamine synthesis from human BM cells. In the present study, we examined the effect of sCD23 on histamine synthesis by human BM cells and its relationship with IL-3. Our data showed that recombinant 25 KDa sCD23, although on its own failed to induce histamine release, significantly enhanced IL-3-stimulated histamine synthesis by BM cells. These data provide further support for the cytokine-like role of sCD23 on hematopoietic cells.

Human thymic epithelial cells produce interleukin-3.

Interleukin-3 (IL-3) is a hematopoietic growth factor suggested to be produced by activated T lymphocytes. Meanwhile, supernatants from human thymic stroma could promote the proliferation of myeloid stem cells. Thus, we investigated whether IL-3 accounts for this activity. Therefore, human thymic epithelial cells (TEC), fibroblasts, and adherent cells were isolated, and their culture supernatants assayed for myeloid colony promotion. Only supernatants from thymic epithelial cells supported colony-forming unit growth in semisolid media. This effect decreased following anti-IL-3 monoclonal antibody addition to these cultures. Furthermore, in situ hybridization showed the presence of IL-3 mRNA in epithelial cells. Effect of TEC culture conditions on IL-3 production by these cells was also studied. Together, these data show that IL-3 production is not the exclusive property of human activated T lymphocytes.

In vitro differentiation and proliferation of purified human thymic and bone marrow CD7+CD2- T-cell precursors.

It is well established that CD7, gp40 antigen is one of the first antigens detected on the surfaces of cells of the human T-cell lineage. Using complement-dependent cytotoxicity and immunoadherence to anti-CD7-coated surfaces, we were able to purify CD7+2-3-4-8-TcR- cells with greater than 90% purity from both human thymus and bone marrow. Limiting dilution analysis showed that these cells displayed high ability to generate mature T-cell clones when they were cultured in the appropriate conditions. These precursors needed phytohemagglutinin-stimulated leukocyte-conditioned medium (PHA-LCM) as a differentiation signal before being able to respond to PHA and recombinant interleukin 2 (rIL2). CD7+CD2- precursors differed from more mature CD7+CD2+ thymocytes because they were not sensitive to PHA, IL2, or CD2 triggering. Bone marrow-derived clones were mostly CD4+, whereas thymic cells generated more CD8+ than CD4+ clones. Together, this study indicates that the CD7+CD2- precursor is one of the earliest prothymocytes able to differentiate and proliferate in vitro.