Journal Logo

Original Articles: Immunobiology and Genomics

Rabbit Antithymocyte Globulin Inhibits Monocyte-Derived Dendritic Cells Maturation In Vitro and Polarizes Monocyte-Derived Dendritic Cells Towards Tolerogenic Dendritic Cells Expressing Indoleamine 2,3-dioxygenase

Gillet-Hladky, Stéphanie1,3; de Carvalho, Camila Miranda1; Bernaud, Janine1; Bendahou, Christine2; Bloy, Christian2; Rigal, Dominique1

Author Information
doi: 10.1097/


Rabbit antithymocyte globulin (rATG) is a preparation containing anti thymocyte polyclonal antibodies, obtained after immunization of rabbits with human thymocytes. Some semiquantitative analyses of rATG have shown that it contains antibodies against many molecules including anti CD6, CD16, CD18, CD28, CD38, CD40, and CD58 antibodies. It has been shown that antibodies against CD3, CD4, CD8, CD11a, CD40, CD45, CD54 and Class II major histocompatibility complex (MHC), which are also present in rATG, persist for a long time in vivo (1). Michallet et al. have shown that rATG contain functional antibodies against βintegrin CD49d/CD29, alpha4beta7 integrin, CD50, CD102 and that rATG binds CCR7 on lymphocytes, CCR5 on monocytes, and CXCR4 on both cells (2).

rATG is used in transplantation because of its strong immunosuppressive properties. These effects were first demonstrated in vitro (3) and in rat skin allograft models (4). The first clinical studies in man were carried out in the 1970s, in cardiac and renal transplantation (5). In kidney transplantation, at that time, rATG use was associated with excellent patient and graft survival rates (6). rATG is also used as second-line treatment of acute rejection (where steroids have failed) in kidney transplantation (7). In heart transplantation, it has been shown that prophylactic use of rATG reduces the incidence of rejection and also the requirement for steroids and other immunosuppressive agents (8). Moreover, rATG is used in combination with corticosteroids for therapy of acute graft versus host disease (GvHD) (9, 10) and aplastic anemia (11). More recently, rATG has been used as part of myeloablative conditioning regimens before bone marrow or allogeneic stem cell transplantation, in order to prevent GvHD (acute and chronic) as well as to assist engraftment (12, 13). rATG has also been used in immunosuppressant regimens aimed at inducing immunological tolerance in kidney (14, 15), heart (16), and hepatic (17) transplantation.

The immunosuppressive mechanisms of action of rATG have been widely studied. It has been shown that rATG induces tolerance that is reproducible and dose dependent in rats (18). Several mechanisms of action of rATG on lymphocytes have been described: rATG induces T cell depletion and impairs the function of T lymphocytes which escape deletion. Moreover, rATG interferes with leukocyte chemotactic responses. rATG is also an inhibitor of integrins that are essential for cellular adhesion and leukocyte-endothelial interactions (2). Recently, another study showed that rATG has also strong effects on B lymphocytes by inducing B lymphocyte apoptosis in vitro (19).

Dendritic cells (DCs) are the main antigen presenting cells; they act both in priming and stimulating immune responses. Overall, they play a central role in the immune system as they can act as stimulators or inhibitors of immune response (20, 21). DCs are a very heterogeneous population that can display many properties depending on their state of maturation and on their phenotype. Immature dendritic cells play the role of sentinels of the immune system. They are present in large numbers where pathogens can enter, existing as Langerhans cells in the skin, lung dendritic cells and Peyer’s patch dendritic cells in the gut (22–24). They show a high ability to internalize and process antigens. Antigen contact induces dendritic cell maturation and migration towards secondary lymphoid organs where they present antigen to effector T cells. DC maturation is accompanied by morphological and phenotypic changes. Mature DCs show larger dendrites that allow greater contact with T cells during stimulation. They show up-regulation of Class II MHC, CD40, CD86, CD80, CD83 and DC-LAMP molecules and down-regulation of the capacity to uptake and process antigen (25, 26). Different types of DCs have been described in vivo: namely DC1 and DC2 subtypes that stimulate TH1 and TH2 pathways respectively of immune responses. Some dendritic cells with tolerogenic properties have been reported (27, 28). Immature dendritic cells have been shown to have tolerogenic properties which probably contribute to peripheral tolerance (29). Among them, the plasmacytoid DCs (PDCs) have a different morphology from the other DCs called myeloid dendritic cells. PDCs do not have dendrites, produce large amounts of type I interferons in response to virus and resemble lymphocytes (30). They display CD123, ILT3 and ILT4 molecules (31) and recently in vivo some tolerogenic dendritic cells have been found to express CCR6 (32). Tolerogenic DCs can act in different ways: by molecular contacts with lymphocytes (ILT3). Some cells express enzymes, such as indoleamine 2–3 dioxygenase (IDO) which metabolizes tryptophan contained in the medium. Tryptophan is an essential amino-acid and its absence prevents lymphocytes from proliferation (33). Finally, production of suppressive cytokines like IL-10, TGF-β or IFN alpha can also induce an immune tolerant state.

The present study examined the action of rATG in vitro on the differentiation and the maturation of Mo-DCs. Firstly, rATG induced a blockage of maturation of Im-Mo-DCs. Secondly, rATG induced the production of a new DC subset (named ATG-DCs) having a tolerogenic pattern. The ATG-DCs were stained by anti-CD123, ILT-3, CCR6 and produced IFNα and IDO mRNA.


Reagents for Cell Culture

RPMI 1640 (Life Technologies, Eggenstein, Germany), which already contains 0.8 mg/ml NaHCO3, was supplemented with glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), and NaHCO3 (1.5 mg/ml). The serum source was a pool of human AB serum (HS), heat inactivated, and added at 10% final concentration.

Recombinant human Granulocyte Macrophage-Colony Stimulating Factor (GM-CSF), Interleukin-4 (IL-4) and Tumor Necrosis Factor-alpha (TNF-α) were purchased from R&D Systems (Abingdon, UK). Lipopolysaccharide (LPS) was obtained from Sigma Chemical (St Quentin Fallavier, France). CD40L-transfected cell lines were kindly provided by S. Lebecque (Schering-Plough, Lyon, France). Polyinosine-polycytidylic acid (poly (I:C)) is a synthetic analogue of double-stranded RNA (dsRNA), with a molecular pattern associated with viral infection. Poly (I:C) was provided by Sigma Chemical Co. (St. Quentin Fallavier, France). rATG and non immune rabbit immunoglobulines (rabbit Ig) were provided by Genzyme Polyclonals SAS (Lyon, France).

Preparation of Monocytes from Peripheral Blood

Fresh peripheral blood was collected from normal healthy volunteers. The whole blood was centrifuged 20 min at 200 g in order to minimize contamination of peripheral blood mononuclear cells (PBMC) with platelets. The supernatant containing most of the platelets and plasma was carefully removed before isolation of PBMCs by centrifugation for 25 min at 600 g in lymphocyte separation medium (Eurobio, Les Ulis, France). After two washes with phosphate buffer saline medium (PBS), a two-step discontinuous density gradient was performed in order to isolate monocytes. With this aim, 50% (6 ml) and 40% (3 ml) dilutions of stock iso-osmotic solution of Percoll (1.130 g/ml; Pharmacia LKB, Uppsala, Sweden) in Dulbecco’s calcium- and magnesium-free PBS containing 5% HS were layered sequentially. 40×106 PBMC suspended in the Dulbecco’s PBS/HS were layered over the gradient and centrifuged at 1000 g for 25 min at 4°C. Low density cells, mostly monocytes, were collected from the interface of the two Percoll solutions and washed twice with PBS. They were then re-suspended in PBS containing 10% HS in a volume of 60 μl per 107 total cells. Monocytes were ultimately isolated by negative selection, using a cocktail of hapten-conjugated CD3, CD7, CD19, CD45RA, CD56, anti-IgE antibodies and MACS microbeads coupled to the anti-hapten monoclonal antibody (Miltenyi Biotec, Paris, France). These magnetically labeled cells were depleted by retaining them on a MACS column in the magnetic field of the MidiMACS, whereas monocytes were eluted from the column by several washes. The purity of monocyte fraction was at least 90%.

Preparation of CD4+ T lymphocytes

High-density cells, mostly lymphocytes, were collected from the pellet of the Percoll solutions and washed twice with PBS. They were removed from the CD14+ cells with CD14 microbeads (from Miltenyi Biotec, Paris, France). The purity of the lymphocytes was superior to 95%. To purify CD4+T lymphocytes fraction, Dynabeads were used, following the protocol given by the supplier (Dynal Biotech, Oslo, Norway). We obtained CD4+T lymphocytes of at least 98% purity.

Generation of Dendritic Cells from Peripheral Blood Monocytes

Duperrier et al. described a method to generate monocyte-derived dendritic cells (34). Monocytes (5×106) were plated on tissue culture wells at 37°C in humidified 5% CO2 in air. The culture medium was supplemented with 200 U/ml rhGM-CSF and 500 U/ml rh-IL4, in a 6 ml final volume. Cells were cultured with or without 100 μg/ml rATG or with rabbit Ig as negative control. On days two and five, removing 3 ml medium and adding back 3 ml of fresh medium with cytokine-fed cultures and rATG or rabbit Ig.

On day six, nonadherent cells were transferred into Teflon beakers. Cells were then cultured at a density of 1×106 cells per well in a 6 ml final volume in the presence of either rhTNFα (recombinant human TNFα) (200 U/ml), CD40L cells (5.105 irradiated CD40L transfected cells for 106 DCs), LPS (1 μg/ml) or Poly(I:C) (12.5 μg/ml) for two additional days. Cells that were in presence of rATG or rabbit Ig at the beginning of the culture were put in presence of rATG or rabbit Ig (100 μg/ml) during maturation.

Antibodies and Flow Cytometric Analysis

Cell staining was performed using FITC- or PE- conjugated and affinity purified mouse monoclonal antibodies (mAb) or unconjugated antibodies with FITC-conjugated sheep anti-mouse immunoglobulin F(ab′)2 fragments (Silenus, Eurobio) for indirect staining. Analyses were performed using a FACS Scan flow cytometer (Becton Dickinson, Franklin Lakes, USA). The mAbs IgG1, IgG2a, anti-CD14, anti-HLA DR, anti-Class I MHC, anti-MR, anti-CD123 and anti-CCR6 were purchased from Becton Dickinson. The mAbs anti-CD83, anti-CD54, anti-CD80, anti-CD86, anti-B7-H1, anti-CCR7 and anti-DC-LAMP were purchased from Immunotech. For intra-cellular DC-LAMP staining, cells were permeabilized with 0.25% saponin (Sigma, Germany) and 4% BSA for 1 hr after fixation in 4% paraformaldehyde for 10 min. Cells were incubated 30 min either with anti-DC-LAMP or control mAb, then washed twice with 0.1% saponin. Annexin V and propidium iodure stainings were performed as described by the provider.

FITC-Dextran Endocytosis

105 immature DCs were cultured at 37°C or 0°C (as control) for 15 min. They were then incubated with 1 mg/ml of FITC-Dextran at 37°C or 0°C for 30 min. The reaction was stopped with cold PBS. Cells were washed three times before being analyzed on a FACS Scan flow cytometer. FITC-Dextran was obtained from Molecular probes (Eugen, USA).

Antigen Specific Presentation by Dendritic Cells and Proliferation Assays

Mature DCs stimulatory function was assessed by incubating 105 autologous or allogeneic CD4+ T lymphocytes with graded numbers of irradiated DCs (30 Gray) or PBMCs (as control) in U-bottom microwells. Graded numbers of these cells were cultured in presence of 105 autologous or allogeneic mononuclear cells in U-bottom microwells. Cell proliferation was measured by [3H]-Thymidine (TdP) incorporation on day four of culture.

Cytokine Detection

Cytokine production during dendritic cell maturation was measured in the supernatant of respectively Mo-DCs by ELISA using matched paired antibodies specific for IL-12 p70 and IL-10 (Becton Dickinson, USA) or IFN alpha (Biosource, Camarillo, USA).

Chemotaxis Assays

We evaluated cell migration in response to several chemokines: RANTES/CCL5 (regulated on activation normal T-cell expressed and secreted) and Macrophage inflammatory protein MIP-3α/CCL20 (from R&D Systems) using a technique reported by Dieu et al. (34). The chemoattractant effect of the chemokines were tested at different concentrations ranging from 1 to 1,000 ng/ml in RPMI 1640 1% HS and added to the lower wells of the chemotaxis chamber in 48-well polycarbonate membrane plates (Costar, Corning, NY). Then, 3×105 cells/well in 100 μL of RPMI 1640 1% HS were added to the upper wells of the chamber, with a standard 5-μm pore polyvinylpyrrolidone-free polycarbonate filter separating the lower wells. Cells were incubated at 37°C in humidified air with 5% carbon dioxide for one hr 30 min. Then, cells that had migrated to the underside of the filter were collected and resuspended in 300 μL of cold 2% paraformaldehyde. The FACScan flow cytometer was then used to determine the cell density of the migrated DCs by an acquisition of 250 μL final volume.

Analysis of Expression of Indoleamine 2,3-dioxygenase by Reverse Transcription

ATG-DCs and Mo-DCs cultured as described were harvested on day 6. Total RNA from the DCs was prepared with Promega (Madison, WI) RNAgents Total RNA Isolation System, as described by provider. Total RNA was reverse transcribed with Reverse transcription System (Promega) and used as template for PCR amplifications. Primers for IDO amplification were ACA GAC CAC AAG TCA CAG CG (sense) and AAC TGA GCA GCA TGT CCT CC (antisense) as described by Nagineni et al. (35). Beta2-microglobulin primers were TTA GCT GTG CTC GCG CTA CTC TCT (sense) and TGT CGG ATT GAT GAA ACC CAG AGA (antisense). The PCR mixture (20 μL) consisted in PCR Reaction Buffer (containing 10 mM Tris-HCl, 50 mM KCl and 1.5 mM MgCl2) (Roche, Basel, Switzerland), a cDNA preparation (corresponding to 50 ng of RNA), 1 μM each primer, 200 μM each of the deoxynucleotide triphosphates and 2.5 U of Taq Polymerase (Roche). PCR amplification was performed on the mixture with a first step at 95°C for five min, then 35 cycles at 95°C for 30s, 57°C for 30s and 72°C for one min. The reaction mixture was kept at 72°C for five min after the PCR cycles were completed. PCR amplification products were visualized on a 2% agarose gel electrophoresis, stained with ethidium bromide and photographed under UV light. The 662 pb products corresponded to IDO cDNA amplification.

Statistical Analysis

Every experiment was performed at least three times to confirm reproductibility of the results. Student’s t test was used to examine the significance of the data. The difference was considered to be statistically significant when P≤0.05.


rATG Inhibits In Vitro Maturation of Immature MoDCs

To study the influence of rATG on dendritic cell maturation, MoDCs were generated as described in materials and methods. Briefly, human monocytes were cultured for six days in the presence of IL-4 and GM-CSF to produce immature dendritic cells (Im-MoDCs) both with and without rATG or with rabbit Ig. On day six, these cells were transferred to Teflon vials for two days in presence of different maturation inducers: TNFα, LPS, CD40L, or Poly (I:C) both with and without rATG or with rabbit Ig. Then, these cells were stained on day eight by using a set of monoclonal antibodies specific for mature dendritic cells: anti-HLA-DR, CD83, CD80, CD86, MR, ILT3, B7-H1, CCR7 and DC-LAMP. Mean and standard deviation (SD) of expression percentages are indicated for each molecule. First, TNFα matured Mo-DCs in presence of rabbit Ig showed similar phenotype to TNFα matured Mo-DCs alone (data not shown). As shown in Figure 1A, TNFα matured MoDCs in presence of ATG showed lower percentages of HLA-DR+, CD86+, CD80+ and MR+ cells. On the other hand, they showed higher expression of ILT-3 and CD123. Results shown in Figure 1A are representative of those obtained with Mo-DCs cultured in presence of CD40-L, LPS and Poly (I:C). Additionally, known dendritic markers, only expressed by mature dendritic cells were studied: CD83, CCR7 and DC-LAMP. In Figure 1B, it can be seen that DC-LAMP is hardly expressed at all by the mature MoDCs cultured in presence of rATG, and that CCR7 and CD83 were weakly expressed by these cells, whatever the maturation agent (TNFα, LPS, CD40L or Poly(I:C) used. In contrast the control study of mature MoDCs without exposure to rATG showed high expression of these molecules. Taken together these results suggest that rATG prevented Im-MoDCs from maturation, whatever the maturation inducers.

(A) Maturation of dendritic cells is impaired in presence of rATG. Monocytes were cultured for six days as described previously. Immature dendritic cells were cultured for two days in presence of TNFα, LPS, CD40L or poly (I:C) with or without rATG. Cells were stained with conjugated MoAbs and analyzed by flow cytometry (gray filled histograms). The black line histograms represent the fluorescence for isotype control MoAbs. Results are shown for TNFα experiments and are representative of those obtained for LPS, CD40L and Poly (I:C) (six independent experiments). Percentages are expressed as the mean±SD of six independent experiments. (B) Expression of CD83, CCR7 and DC-LAMP by matured dendritic cells in presence of rATG. Monocytes were cultured for six days as described previously. Immature dendritic cells were cultured for two days in presence of TNFα, LPS, CD40L or poly (I:C) with or without rATG. Cells were stained with conjugated MoAbs and analyzed by flow cytometry (gray filled histograms). The black line histograms represent the fluorescence for isotype control MoAbs. Percentages are expressed as the mean±SD of six independent experiments.

IL-4+ GM-CSF+ rATG Induced In Vitro a New Subset of Mo-DCs Called ATG-DCs

In a second series of experiments, the effect of the addition of rATG on the production of Im-MoDCs induced by IL4+ GM-CSF was examined. Monocytes were cultured for six days in the presence of IL-4 and GM-CSF, with (ATG-DCs) or without rATG (Im-MoDCS) or with rabbit Ig.

They were stained with monoclonal antibodies specific for HLA-DR, CD83, CD14, CD80, CD86, CD40, CD54, ILT3, B7-H1, CCR7 and DC-LAMP molecules. Results shown in Figure 2A are representative of ten independant experiments. Mean and SD of expression percentages are indicated for each molecule. First, we can observe that Mo-DCs + rabbit Ig are similar to Mo-DCs, suggesting a specific effect of rATG. ATG-DCs exhibit a phenotype very close to the one of Im-MoDCs with a weak expression of CD83, DC-LAMP and CCR7 molecules (data not shown). The percentages of positive cells for HLA-DR, CD86, CD14 are similar for the two types of cells. Additionally, ATG-DCs showed a lower percentage of CD54 positive cells. However, as shown in Figure 2B, ATG-DCs expressed the same level of MR molecules as Im-MoDCs. But their ability to phagocytose dextran-FITC molecules was very weak. These results suggested that ATG-DCs are not yet able to internalize antigen and so to present it. A final result is that ATG-DCs do not undergo apoptosis as shown by Propidium Iodure and Annexin V staining performed on day six and eight. These results are shown in Figure 2C.

(A) Differentiation of monocytes into dendritic cells in presence of rATG. Monocytes were cultured for six days as described previously with non specific rabbit immunoglobulins or without (Mo-DCs) or with ATG (ATG-DCs) or in presence of rabbit Ig. Cells were stained with conjugated MoAbs and analyzed by flow cytometry (gray filled histograms). The black line histograms represent the fluorescence for isotype control MoAbs. Percentages are expressed as the mean±SD of 10 independent experiments. (B) FITC-Dextran internalization by ATG-DCs. Mo-DCs and ATG-DCs were tested for their ability to internalize the FITC-Dextran (gray filled histograms). As a control, the black line histograms represent the fluorescence values of FITC-Dextran uptake at 0°C. Cells were stained with conjugated MoAbs specific for Mannose receptor and analyzed by flow cytometry (gray filled histograms). The black line histograms represent the fluorescence for isotype control MoAbs. Percentages are expressed as the mean±SD of three independent experiments. (C) rATG do not induce apoptosis of dendritic cells. Mo-DCs and ATG-DCs and ATG-DCs+TNFα were stained with Annexin V mAb and propidium ioduria as described previously.

ATG-DCs Express Tolerogenic Markers ILT3, CD123, and CCR6

Interestingly, ATG-DCs showed an up-regulation of ILT3 and CD123 molecule expression. In Figure 3A, CD123 expression is shown according to the rATG concentration in the culture. Rabbit IgG was used as negative control. Means of expression percentages and SD are noted for each histogram. CD123 expression level parallels rATG concentration. Immunofluorescent staining using anti-ILT3 and anti-CD123 mAbs were performed on ATG-DCs and Mo-DCs as shown in the dot-plot in Figure 3B. Interestingly, CD123 positive ATG-DCs expressed ILT3 molecules. Moreover, Munn et al described in vivo CD123+ CCR6+ DCs with tolerogenic properties (32). Immunofluorescent stainings for CCR6 were performed on ATG-DCs and Mo-DCs. In Figure 4A, it can be seen that ATG-DCs expressed CCR6, whereas this receptor was weakly expressed by Mo-DCs.

(A) Expression of CD123 by ATG-DCs is dose-dependent. ATG-DCs were cultured in presence of different concentrations of ATG ranging from 0 to 200 μg/ml (blue line histograms). Nonspecific rabbit Ig were used as external control. The red line histograms represent the fluorescence for isotype control MoAbs. Percentages are expressed as the mean±SD of three independent experiments. (B) ATG-Dcs are CD123+ILT3+. Monocytes were cultured for six days as described previously both without (Mo-DCs) or with or ATG (ATG-DCs). Cells were stained with conjugated MoAbs and analyzed by flow cytometry. Percentages are expressed as the mean±SD of six independent experiments.
ATG-DCs express functional CCR6. Monocytes were cultured for six days as described previously both without (Mo-DCs) or with orATG (ATG-DCs). (A) Cells were stained with conjugated anti-CCR6 MoAbs and analyzed by flow cytometry. (gray filled histograms). The black line histograms represent the fluorescence for isotype control MoAbs. Percentages are expressed as the mean±SD of six independent experiments. (B) ATG-DCs (black histograms) and Mo-DCs (gray histograms) were tested for their response to RANTES/CCL5 (1–1000 ng/ml) or MIP3α/CCL20 (1–1000 ng/m). The optimal concentrations are shown. The results are expressed as the ratio of migrating cells in the presence of chemokine or medium alone. Data shown are expressed as the mean±SD of three independent experiments. (C) Cytokine secretion by ATG-DCs. Supernatants of ATG-DCs (black histograms) and Mo-DCs (gray histograms) were harvested after six days of culture Using ELISA test, presence of IL-10, IL-12 and IFNα were researched. Results shown are mean±SD of three separate experiments.

ATG-DCs Respond to MIP3α/CCL20 Chemoattractant

CCR6 is a receptor for MIP3α/CCL20 chemokine. Thus, migration properties of ATG-DCs were studied to evaluate the functionality of this receptor. In Figure 4B, the chemotactic effect of RANTES/CCL5, and MIP3α/CCL20 was evaluated. The effect of MIP3β/CCL19 and 6Ckine/CCL21 was not studied because Im-MoDCs or ATG-DCs didn’t express CCR7 (the receptor for these two chemokines) (data not shown). Mo-DCs migrate towards RANTES/CCL5 gradient and were not sensitive to MIP3α/CCL20. ATG-DCs reacted more weakly to the chemoattractant effect of RANTES/CCL5 than Mo-DCs. By contrast, ATG-DCs were chemoattracted by MIP3α/CCL20 showing that ATG-DCs expressed functional CCR6 chemokine receptors.

ATG-DCs Produce IFNα

As rATG induced DCs with tolerogenic features or plasmacytoid like cells as shown by their phenotype, a study of their cytokine secretion was performed. IL-12, IL-10 and IFNα secretion was studied as shown in Figure 4C. Im-MoDCs and ATG-DCs did not produce significant amounts of IL-12 and IL-10. Im-MoDCs showed no IFNα secretion whereas ATG-DCs were found to secrete some IFNα. This finding suggested that ATG-DCs would prime preferentially in the TH2 way and would be suppressive for the TH1 immune response. Interestingly, IFNα secretion was studied in ATG-DCs exposed to TNFα and these cells were found to secrete this cytokine.

ATG-DCs Express 2–3 Indoleamine Dioxygenase mRNA

ATG-DCs were shown to express ILT3, CD123, and CCR6 molecules and produce some IFNα. This pattern suggested strongly that ATG-DCs belong to the tolerogenic DCs subset. The expression of Indoleamine 2, 3-dioxygenase (IDO) in the ATG-DCs was studied by testing IDO gene expression. Total RNAs of Im-MoDCs and ATG-DCs were isolated. A random primers reverse transcription was then performed. PCR amplification was done on the reverse transcribed products with IDO and β2m primers (as described previously). PCR products were visualized in agarose gel electrophoresis after being photographed. As shown in Figure 5, amplification band of the waited size for IDO cDNAs could be observed for ATG-DCs and not for Mo-DCs. These results showed that ATG-DCs expressed IDO mRNA and suggest strongly that these cells produced IDO and then were able to inhibit T lymphocyte proliferation.

ATG-DCs express mRNA for IDO. After six days of culture, Mo-DCs and ATG-DCs were harvested for RNA purification. RT-PCR was performed to test the presence of IDO mRNA as described in Materials and Methods.

ATG-DCs Inhibit In Vitro Autologous and Allogeneic MLR

The ability of mature dendritic cells cultured in the presence or not of rATG was tested first by allogeneic and autologous mixed leukocyte reaction (MLR). Graded number of dendritic cells were put in presence of stable number of autologous or allogeneic CD4+ T lymphocytes (purity superior to 95%). Results are shown for TNFα exposed ATG-DCs and are representative of those obtained with LPS, CD40L and Poly (I:C). Figure 6 shows that the allogeneic reaction was strongly disturbed when matured ATG-DCs were used. Moreover, the autologous reaction was very weak, compared to the one obtained with matured Mo-DCs. These results were consistent with the phenotype of the dendritic cells cultured in presence of rATG previously described.

ATG-DCs impaired autologous and allogenic MLR. Mo-DCs cultured in the presence of TNFα without (gray histograms) or with rATG (black histograms) in autologous (A) or allogenic (B) MLR. Graded number of dendritic cells (102 to 105) was used as stimulator cells and either autologous or allogenic CD4+ T lymphocytes (105) were used as responders. Cell proliferation was measured by [3H]-TdP incorporation. (Data are reported as mean values±SD.) Results shown are representative of three separate experiments.


rATG are used in conditioning treatment for bone marrow transplantation, in the treatment of acute graft versus host disease, in the prevention or treatment of acute rejection in organ transplantation and in severe bone marrow aplasia. rATG effects on lymphocytes have been widely studied on peripheral blood lymphocytes in vivo and in vitro. First, at low doses, ATG has mitogenic properties: by activating blastogenesis of CD4+, CD8+ and CD56+ (NK) lymphocytes. It induces also IL-2 and IFN α secretion (36). But rATG are also used for their tolerogenic properties: they induce lymphocyte depletion but also have functional effects on lymphocytes. Bonnefoy-Berard et al. showed that inhibition of T cell proliferation by ATG can be attributed to a posttranscriptional defect of CD25 expression (37). Genestier et al. have shown in vitro that at high concentrations, rATG activates the human classic complement pathway and induces lysis of both resting and PHA-activated peripheral blood mononuclear cells (38). rATG also triggers surface Fas expression in naive T cells and Fas-ligand gene and protein expression in both naive and primed T cells resulting in Fas/Fas-L interaction mediated cell death (38). In vivo, rATG produces rapid, dose dependent T cell depletion in peripheral lymphoid tissues where apoptotic cells can be demonstrated in T cell zones. There may be a role for cathepsin B in T cell apoptosis at concentrations used in clinical treatment (39).

Up to now, no work has examined the effect of rATG on dendritic cells even though these cells play a central role in the immune system. Firstly, rabbit Ig has no relevant effect on differenciation or maturation of Mo-DCs, suggesting a specific efffect of rATG. We found that rATG prevents immature Mo-DCs from maturation. In presence of rATG, immature dendritic cells showed an immature phenotype with a low expression of CD83, DC-LAMP or CCR7 molecules which are maturation markers. Moreover, in presence of rATG, immature Mo-DCs exhibited a specific phenotype with an up-regulation of ITL3 in a dose dependent manner. Similarly, up-regulation of CD123 was found and all cells that were CD123+ were also ILT3+. Importantly, in contrast to the effect on lymphocytes, rATG did not induce apoptosis of Mo-DCs. ATG-DCs expressed ILT3 molecules, resembling some tolerogenic dendritic cells described in vivo. Interestingly, ATG-DCs expressed CCR6 but not CCR7 or CCR5 (data not shown). This receptor is functional at the surface of ATG-DCs as they are sensitive to the chemoattractant effect of MIP3α/CCL20. Also, the cytokine expression pattern was studied for ATG-DCs. These cells secreted IFNα suggesting that ATG-DCs would prime in the TH2 direction and would be suppressive for the TH1 immune response. Many studies have shown that ATG prevents graft versus host disease: during GvHD, cytotoxic mechanisms are activated. Interestingly, Suciu-Focca et al. demonstrated that IFN-α treatment of Mo-DCs induced strong upregulation of ILT3 molecules (31). Moreover, in this study, ATG-DCs were found to express mRNA of IDO. This enzyme has been shown to have a strong role in the catabolism of tryptophan. Cells expressing IDO can suppress T-cell responses and promote tolerance (40).

In vivo, tolerogenic dendritic cells constitute a heterogeneous subset. PDCs morphology is similar to plasma cells, as they show no dendrites. They express CD123, ILT3, BDCA-2, BDCA-4, CD4, and CD62L, have no myeloid markers and secrete high amounts of IFN-α. Other tolerogenic DCs have been described. Munn et al. showed that circulating tolerogenic DCs expressed CD123 and CCR6 (32). Moreover, expression of ILT3 and ILT4 and expression of IDO are essential features of tolerogenic dendritic cells (31).

Altogether, our results show that rATG permits the in vitro generation of a new dendritic cell population with many properties of tolerogenic dendritic cells. ATG-DCs express CD123, ILT3, CCR6 but not BDCA-2, BDCA-4. Phenotypically, they resemble strongly those described in vivo by Munn et al. (32). They produce some IFNα like in vivo PDCs. Functionally; they have two main characters of tolerogenic dendritic cells: ILT-3 and IDO expression. ILT-3 has been described as a main marker of tolerogenic dendritic cells by Suciu-focca et al. Expression of IDO is also strong evidence that suggests tolerogenic properties for ATG-DCs. These cells are very interesting because they are consistent with the concept of acquired tolerance in vivo. As the depletion of lymphocytes by ATG is transient and stopped when administration of ATG stopped, tolerogenic dendritic cells induced by ATG are permanent and tolerance can be maintained by these cells even after treatment had stopped.


1. Rebellato LM, Gross U, Verbanac KM, Thomas JM. A comprehensive definition of the major antibody specificities in polyclonal rabbit antithymocyte globulin. Transplantation 1994; 57(5): 685.
2. Michallet MC, Preville X, Flacher M, et al. Functional antibodies to leukocyte adhesion molecules in antithymocyte globulins. Transplantation 2003; 75(5): 657.
3. Woodruff M. Effect of antilymphocytic serum on suspensions of lymphocytes in vitro. Nature 1951; 168(4262): 35.
4. Woodruff M. Effect of lymphocyte depletion by thoracis duct fistula and administration of antilymphocytic serum on the survival of skin homografts in rats. Nature 1963; 200: 702.
5. Lance E. Induction of tolerance with antilymphocytic serum transpl. Proc 1969; 1(1): 429.
6. Filo R, Smith E. Therapy of acute cadaveric renal allograft rejection with adjunctive antithymocyte globulin. Transplantation 1980; 30(6): 445.
7. Hoitsma A, Reekers P, Kreeftenberg J, et al. Treatment of acute rejection of cadaveric renal allografts with rabbit antithymocyte globulin. Transplantation 1982; 33(1): 12.
8. Kawaguchi A, Szentpetery S, Mohanakumar T, et al. Effects of prophylactic rabbit antithymocyte globulin in cardiac allografts recipients treated with cyclosporine. J Heart Transplant 1987; 6(4): 214.
9. Dugan MJ, DeFor TE, Steinbuch M, et al. ATG plus corticosteroid therapy for acute graft-versus-host disease: predictors of response and survival. Ann Hematol 1997; 75(1–2): 41.
10. Remberger M, Aschan J, Barkholt L, Tollemar J, Ringden O. Treatment of severe acute graft-versus-host disease with anti-thymocyte globulin. Clin Transplant 2001; 15(3): 147.
11. Teramura M, Kobayashi S, Iwabe K, Yoshinaga K. H M Mechanism of action of antithymocyte globulin in the treatment of aplastic anaemia: in vitro evidence for the prensence of immunosuppressive mechanism. Br J Haematol 1997; 96(1): 80.
12. Kang H, Shin H, Choi H. Fludarabine, cyclophosphamide plus thymoglobulin conditioning regimen for unrelated bone marrow transplantation in severe aplastic anemia. Bone Marrow Transplant 2004; 34(11): 939.
13. Mohty M, Boiron J, Damaj G, et al. Graft-versus-myeloma effect following antithymocyte globulin-based reduced intensity conditioning allogeneic stem cell transplantation. Bone Marrow Transplant 2004; 34(1): 77.
14. Shapiro R, Jordan ML, Basu A, et al. Kidney transplantation under a tolerogenic regimen of recipient pretreatment and low-dose postoperative immunosuppression with subsequent weaning. Ann Surg 2003; 238(4): 520.
15. Buchler M, Hurault de Ligny B, et al. Induction therapy by anti-thymocyte globulin (rabbit) in renal transplantation: a 1-yr follow-up of safety and efficacy. Clin Transplant 2003; 17(6): 539.
16. Carlsen J, Johansen M, Boesgaard S, et al. Induction therapy after cardiac transplantation: A comparison of anti-thymocyte globulin and daclizumab in the prevention of acute rejection. J Heart Lung Transplant 2005; 24(3): 296.
17. Bonatti H, Mühlmann G, Labeck B, et al. Effects of ATGinduction on laboratory parameters, early complications and outcome after liver transplantation. Transplantation 1999; 67(9): S633.
18. Mjornstedt L, Olausson M, Hedman L, et al. Induction of long-term heart allograft survival in the rat by rabbit ATG. Int Arch Allergy Appl Immunol 1984; 74(3): 193.
19. Zand MS, Ifthikharuddin JJ, Abboud C. Induction of myeloma cell apoptosis by polyclonal rabbit anti-thymocyte globulin (rATG). J Clin Oncol 2004; 22(14S): 2552.
20. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998; 392(6673): 245.
21. Sallusto F, Lanzavecchia A. Mobilizing dendritic cells for tolerance, priming, and chronic inflammation. J Exp Med 1999; 189(4): 611.
22. Inaba K, Schuler G, Witmer MD, et al. Immunologic properties of purified epidermal Langerhans cells. Distinct requirements for stimulation of unprimed and sensitized T lymphocytes. J Exp Med 1986; 164(2): 605.
23. Cochand L, Isler P, Songeon F, Nicod LP. Human lung dendritic cells have an immature phenotype with efficient mannose receptors. Am J Respir Cell Mol Biol 1999; 21(5): 547.
24. Brandtzaeg P. Development and basic mechanisms of human gut immunity. Nutr Rev 1998; 56: S5.
25. Zhou LJ, Tedder TF. Human blood dendritic cells selectively express CD83, a member of the immunoglobulin superfamily. J Immunol 1995; 154(8): 3821.
26. Lechmann M, Zinser E, Golka A, Steinkasserer A. Role of CD83 in the immunomodulation of dendritic cells. Int Arch Allergy Immunol 2002; 129(2): 113.
27. O'Doherty U, Peng M, Gezelter S, et al. Human blood contains two subsets of dendritic cells, one immunologically mature and the other immature. Immunology 1994; 82(3): 487.
28. Robinson S, Patterson S, English N, et al. Human peripheral blood contains two distinct lineages of dendritic cells. Eur J Immunol 1999; 29(9): 2769.
29. Steinman RM. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. PNAS 2001; 99(1): 351.
30. Colonna M, Trinchieri G. Plasmacytoid dendritic cells in immunity. Nat Immunol 2004; 5(12): 1219.
31. Manavalan JS, Rossi PC, Vlad G, et al. High expression of ILT3 and ILT4 is a general feature of tolerogenic dendritic cells. Transpl Immunol 2003; 11(3–4): 245.
32. Munn DH, Sharma MD, Lee JR, et al. Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 2002; 297(5588): 1867.
33. Moffet JR, Ma N. Tryptophan and the immune response. Immunol Cell Biol 2003; 81: 247.
34. Duperrier K, Eljaafari A, Dezutter-Dambuyant C, et al. Distinct subsets of dendritic cells resembling dermal DCs can be generated in vitro from monocytes, in the presence of different serum supplements. J Immunol Methods 2000; 238(1–2): 119.
35. Nagineni CN, Pardhasaradhi K, Martins MC, et al. Mechanisms of interferon-induced inhibition of Toxoplasma gondii replication in human retinal pigment epithelial cells. Infect Immun 1996; 64(10): 4188.
36. Bonnefoy-Berard N, Vincent C, Verrier B. Monocyte-independent T-cell activation by polyclonal antithymocyte globulins. Cell Immunol 1992; 143(2): 272.
37. Bonnefoy-Berard N, Verrier B, Vincent C. Inhibition of CD25 (ILR2α) expression and T cell proliferation by polyclonal anti-thymocyte globulins. Immunology 1992; 77(1): 61.
38. Genestier L, Fournel S, Flacher M, et al. Induction of Fas (Apo-1, CD95)-mediated apoptosis of activated lymphocytes by polyclonal antithymocyte globulins. Blood 1998; 91(7): 2360.
39. Michallet MC, Saltel F, Preville X, et al. Cathepsin-B-dependent apoptosis triggered by antithymocyte globulins: a novel mechanism of T-cell depletion. Blood 2003; 102(10): 3719.
40. Mellor AL, Munn DH. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat Rev Immunol 2004; 4(10): 762.

In vitro dendritic cell; Tolerance; Rabbit antithymocyte globulin

© 2006 Lippincott Williams & Wilkins, Inc.