Published online before print June 18, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Immune-Mediated Section, Department of Infectious, Parasitic and Immune-Mediated Diseases, Istituto Superiore di Sanitá, Rome, Italy
3 Correspondence: Immune-Mediated Section, Department of Infectious, Parasitic and Immune-Mediated Diseases, Istituto Superiore di Sanitá, Rome, Italy. E-mail: monica.boirivant{at}iss.it
ABSTRACT
Cholera toxin B subunit (CTB) is an efficient mucosal carrier molecule for the generation of immune responses to linked antigens. There is also good evidence that CTB acts as an immunosuppressant, as it is able to down-modulate human monocyte/macrophage cell line activation and to suppress Th1-type responses. In the present study, we examined the possibility that recombinant CTB (rCTB) may affect human dendritic cell (DC) functions in response to LPS stimulation and may induce the generation of DC with the capacity to generate CD4+ regulatory T cells (Tregs). Our findings show that rCTB partially prevents the LPS-induced maturation process of monocyte-derived DC (MDDC) and decreases their IL-12 production with no relevant effect on IL-10 production. LPS-stimulated MDDC pretreated with rCTB are able to promote the induction of low proliferating T cells, which show an enhanced IL-10 production associated with a reduced IFN-
production and the same high levels of TGF-β as the control. These T cells suppress proliferation of activated autologous T cells. Transwell experiments and blockade of IL-10R and TGF-β showed that the immunomodulatory effect is mediated by soluble factors. Thus, T cells induced by rCTB-conditioned MDDC acquire a regulatory phenotype and activity similar to those described for type 1 Tregs.
Key Words: antigen-presenting cells • cytokines • tolerance
INTRODUCTION
Cholera toxin (CT) is an enterotoxin produced by Vibrio cholerae consisting of a toxic A subunit noncovalently linked with a pentameric B subunit (CTB), which is necessary for binding to cell-surface monosialogangliosides and facilitation of toxin entry. The holotoxin represents the most powerful mucosal adjuvant in animal models of mucosal immunization. As a result of its toxicity, attempts have been made to use its nontoxic B subunit as a carrier for the delivery of the antigen at the mucosal level, which when covalently bound to the B subunit, has been proven to be more potent than the administration of antigen alone for the induction of mucosal antibody responses [1 ] and in animal protocols of mucosal tolerance induction [2 3 4 5 6 ]. Indeed, autoantigen-CTB conjugates have been used in animal models to suppress the autoimmune disease development in experimental allergic encephalomyelitis, spontaneous autoimmune diabetes, experimental autoimmune arthritis, and chondritis [2 3 4 5 6 ].
More recently, immunomodulatory function of unconjugated CTB, in purified and recombinant form, has been explored in in vivo and in vitro studies [7
8
9
10
11
]. Parenteral administration of uncoupled CTB suppresses diabetes development in the NOD mice [9
]. Moreover, oral administration of recombinant (r)CTB inhibits the IL-12-mediated murine experimental colitis induced by intrarectal administration of trinitrobenzene sulfonic acid (TNBS) [10
]. Furthermore, CTB is able to inhibit established Th1-mediated inflammation, as it inhibits IL-12 and IFN- production in ex vivo human Crohn’s disease mucosal explants [11
]. In vitro studies about purified immune cell populations showed that free CTB decreases the expression of CD80 and CD86 on antigen-pulsed murine dendritic cells (DC) [7
]. CTB was also reported to suppress in vitro human macrophage reactivity, inhibiting the proinflammatory response to LPS without affecting the IL-10 production [8
].
DC are professional APCs, which in their immature form, continuously patrol their microenvironment in peripheral tissues for invading pathogens. On pathogen encounter, immature DC develop into mature, immunostimulatory DC. Besides the polarization of conventional Th1 and Th2 cells, DC are able to generate adaptive regulatory T cells (Tregs) [12 ]. In particular, CD4+ type 1 Tregs (Tr1) differentiate in the periphery from naïve precursors stimulated by DC in the presence of IL-10 and ultimately regulate T cell response via their ability to produce IL-10 [13 14 15 ]. A growing body of evidence indicates that adaptive Tregs can develop in the periphery from uncommitted naïve or memory T cells triggered by DC. The current belief is that Treg development is mainly induced by immature or semi-mature DC [16 ].
In the present study, we investigated whether rCTB may affect human DC maturation and functions in response to LPS stimulation and may induce the generation of DC with the capacity to induce CD4+ Tregs. Our findings show that rCTB partially prevents the maturation process of monocyte-derived DC (MDDC) and decreases their IL-12 production upon LPS stimulation. Of great interest, LPS-stimulated MDDC pretreated with rCTB are able to promote the induction of low proliferating T cells, which show an enhanced production of IL-10 and a reduced production of IFN- and have the ability to suppress proliferation of autologous T cells. Thus, T cells induced by rCTB-conditioned MDDC acquire a regulatory phenotype and a functional behavior similar to those described for Tr1 cells.
MATERIALS AND METHODS
Chemical and reagents
The V. cholerae strain 0395-tacCTB, lacking the CTA gene (kindly supplied by Dr. Rino Rappuoli, Istituto Ricerche Immunobiologiche, Chiron, Siena, Italy), was used as a source for rCTB production. rCTB was produced and purified as reported previously [10
], according to the protocol described by Lebens et al. [17
] with minor modifications. All rCTB preparations contained <1 EU/ml endotoxin, as determined by a quantitative, chromogenic Limulus amebocyte lysate test (QLC1000, BioWhittaker, Walkersville, MD, USA). rCTB was dissolved in culture medium and used at 10 µg/ml.
A possible cytotoxic effect of the drugs was excluded using evaluation of cell viability with trypan blue (more than 95% viable cells). Any possible contamination by enterotoxins was excluded by performing control experiments with boiled rCTB.
Human PBMC isolation
Human PBMC were isolated by the Ficoll-Hypaque method from buffy coats from healthy, voluntary blood donors (courtesy of the Transfusional Medicine Department, Policlinico Umberto I, University "La Sapienza," Rome).
Generation and culture of MDDC
Human MDDC were generated according to the protocol described by Sallusto and Lanzavecchia [18
]. Briefly, monocytes were isolated from PBMC by MACS (Miltenyi Biotec, Germany) using anti-CD14 microbeads. The recovered monocytes, which were >95% pure, as shown by flow cytometry with an anti-CD14 fluorescent antibody, were cultured at 5 x 105/ml in RPMI 1640 (BioWhittaker), supplemented with 15% FCS (Hyclone, Logan, UT, USA), 1% L-glutamine, and 1% penicillin/streptomycin (Sigma-Aldrich, St. Louis, MO, USA) containing 50 ng/ml GM-CSF (Peprotech Inc., Rocky Hill, NJ, USA) and 1000 U/ml recombinant human (rh)IL-4 (Endogen, Pierce Biotechnology Inc., Rockford, IL, USA, distributed by Tema Ricerca, Italy) at 37°C and 5% CO2 for 5 days. At Day 5, after monocyte differentiation in immature MDDC (>90% CD1a+CD14– cells, analyzed by flow cytometry), separate cultures were treated or not with 10 µg/ml rCTB for 24 h. Subsequently, cells were harvested, washed extensively, and stimulated with 10 ng/ml LPS from Escherichia coli (Sigma-Aldrich) (LPS-MDDC and CTB-LPS-MDDC) or left untreated (MDDC and CTB-MDDC) for an additional 24 h or 48 h. After this period, MDDC were washed extensively before further use. The phenotype of MDDC, CTB-MDDC, LPS-MDDC, and CTB-LPS-MDDC was analyzed for CD40, HLA-DR, CD80, CD86, and CD83 expression by flow cytometry. At the end of the culture period, cell supernatants were collected and analyzed for IL-10 and IL-12p70 production. A portion of CTB-LPS-MDDC and LPS-MDDC was irradiated (3000 rads) and used to stimulate allogeneic PBMC in MLR.
MLR
PBMC used as responder cells (1x105) were cultured in 96-well plates (Costar Corp., Germany) with graded numbers of irradiated (3000 rads), allogeneic LPS-MDDC or CTB-LPS-MDDC. At Day 4, cells were pulsed with methyl 3H-thymidine (Amersham Life Science, UK) for the last 18 h of culture. 3H-Thymidine incorporation was measured in a LKB Betaplate liquid scintillation counter (Wallac, Inc., UK). Assays were performed in triplicate, and the results were expressed as mean cpm ± SEM.
T cell culture
CD4+CD45RA+ lymphocytes were positively selected from human peripheral blood of healthy donors using the CD4+ Isolation Kit II followed by CD45RA magnetic beads (MACS, Miltenyi Biotec), according to the manufacturer’s instructions. Recovered CD4+CD45RA+ cells were more than 95% pure, as indicated by flow cytometry. Purified T cells were resuspended at the concentration of 1 x 106/ml in RPMI 1640, supplemented with 10% of FCS, 1% L-glutamine, and 1% penicillin/streptomycin (Sigma-Aldrich; complete medium), and cultured with fresh, allogeneic LPS-MDDC or CTB-LPS-MDDC at the ratio of 10:1 for 1 week. After this period of culture, a portion of T cells stimulated with LPS-MDDC (TLPS-MDDC) and T cells stimulated with CTB-LPS-MDDC (TCTB-LPS-MDDC) was harvested, washed, and stimulated with plate-bound, anti-human CD3
(10 µg/ml, R&D Systems Inc., Minneapolis, MN, USA) and soluble anti-human CD28 (2 µg/ml) mAb (R&D Systems Inc.) to analyze their cytokine production (after 24–48 h of culture) and proliferative capacity (after 72 h of culture). Supernatants were collected after 24 h for IFN- and 48 h for IL-10, TGF-β1, and IL-4.
Cytokine assays
Cytokine content in the culture supernatants was measured by sandwich ELISA using commercially available kits: IL-10, IL-12p70, and TGF-β1 ELISA kits (R&D Systems Inc.) and IL-10, IFN-, and IL-4 ELISA kits (Pierce Biotechnology Inc., distributed by Tema Ricerca). ODs were measured on a Bio-Rad Novapath ELISA reader, and the results were expressed as pg/mL ± SEM.
Analysis of T cell function (suppression assay)
Next, we investigated the capacity of TCTB-LPS-MDDC to suppress autologous TLPS-MDDC proliferation. To this purpose, parallel cultures of TLPS-MDDC and TCTB-LPS-MDDC were cultured in the presence of rhIL-2 (40 U/ml, Roche Diagnostics GmbH, Germany). After 6 days, cells were harvested and washed. TLPS-MDDC were then labeled with 5 µM final concentration of the intracellular fluorescent dye CFSE (VybrantTM carboxyfluorescein diacetate, succinimidyl ester cell tracer kit, Molecular Probes, Eugene, OR, USA). To distinguish the two populations, we stained TCTB-LPS-MDDC with PKH-26 dye (2x10–6 M; PKH26 red fluorescent cell linker kit, Sigma-Aldrich). After the cell-labeling, according to the manufacturer’s instructions of each product, cells were resuspended in complete medium and cultured together at different ratios. Results of proliferation and IL-10 production analyses indicated 1:1 as the optimal ratio. To induce activation, T cells were stimulated with LPS-MDDC from the same donor used for the CD4+CD45RA+ T cell priming at a 10:1 ratio for 4 days. In some experiments, 30 µg/ml neutralizing anti-IL-10R mAb (BD PharMingen, San Diego, CA, USA) and 50 µg/ml anti-TGF-β1, -β2, and -β3 mAb (R&D System Inc.) were added at the beginning and at Day 2 of the culture period. The proliferation of the CFSE-labeled TLPS-MDDC was assessed at Day 4.
To assess if cell–cell contact was necessary for the suppressive capacity of TCTB-LPS-MDDC, Transwell experiments were performed using CFSE-labeled TLPS-MDDC cells and PKH-26-labeled TCTB-LPS-MDDC. We placed 4 x 105 CFSE-labeled TLPS-MDDC cells with allogeneic LPS-MDDC (10:1 ratio) in the bottom well of a Transwell system (Costar, Corning Inc., Corning, NY, USA) and 4 x 105 PKH-26-labeled TCTB-LPS-MDDC with allogeneic LPS-MDDC (10:1 ratio) in the upper Transwell chamber. Transwell cultures with 4 x 105 CFSE-labeled TLPS-MDDC cells with allogeneic LPS-MDDC (10:1 ratio) in the upper chamber were performed as control. After 4 days, we measured the proliferative response of CFSE-labeled TLPS-MDDC cells in the bottom well.
Proliferation of CFSE-labeled cells was assessed by a BD FACSCanto flow cytometer and FACSDiva software (BD Bioscience, San Jose, CA, USA). Results were analyzed with the proliferation model of ModFit LT software (Verity Software House, Inc., Topsham, ME, USA) and expressed as proliferation index (P.I.).
Flow cytometry analysis
Cells were washed with PBS and then stained with various fluorochromes using standard methods provided by the manufacturers. Antibodies used for surface staining of human monocytes and DC—PE-labeled anti-CD14 and FITC-labeled anti-CD1a, -CD40, -CD80, -CD83, -CD86, and -HLA-DR mAb—were purchased from BD PharMingen. To analyze the CD4+ T cell population, FITC-labeled anti-CD4, PE-labeled anti-CD45RA, and PE-labeled anti-CD25 mAb (BD PharMingen) were used. For detection of forkhead box p3 (Foxp3)+ cells, T cells were fixed and permeabilized according to the manufacturer’s instructions and incubated with anti-human Foxp3-allophycocyanin mAb (e-Bioscience, San Diego, CA, USA). Isotype-matched antibodies were used as controls, and cells were preincubated with human IgG to avoid nonspecific binding to FcRs. All samples were analyzed using a BD FACSCanto flow cytometer and FACSDiva software (BD Bioscience).
Statistical analysis
Results were expressed as mean ± SEM or as P.I. Statistical analysis was performed using Student’s t-test and Wilcoxon test for paired data as appropriate. P values <0.05 were considered significant.
RESULTS
rCTB treatment affects maturation of human MDDC
We first assessed the influence of rCTB on the pattern of expression of costimulatory (CD40, CD80, CD86), maturation (CD83), and class II MHC (HLA-DR) molecules in MDDC, generated from monocytes cultured with GM-CSF and IL-4. To this purpose, separate cultures of MDDC were treated or not with 10 µg/ml rCTB for 24 h. Subsequently, cells were washed and stimulated with LPS (10 ng/ml; LPS-MDDC and CTB-LPS-MDDC) or left untreated (MDDC and CTB-MDDC). In preliminary experiments, the 24-h and 48-h treatment with LPS on CTB-MDDC and untreated MDDC has been tested. Indeed, we observed that the effect of rCTB on MDDC maturation was more evident during the first 24 h of LPS stimulation. Therefore, in subsequent experiments, we used the 24-h LPS stimulation. As shown in Table 1
, pretreatment with rCTB induces a statistically significant reduction of the LPS-induced up-regulation of CD40 (P=0.037), and it does not affect the expression of the other molecules. rCTB does not induce significant modification in the surface marker expression of MDDC not treated with LPS.
|
View this table: [in a new window] |
Table 1. rCTB Interferes with the Expression of the Costimulatory Molecule CD40 during Human MDDC Maturationa
|
 View larger version (26K): [in a new window] |
Figure 1. rCTB inhibits IL-12p70 production by LPS-MDDC, whereas it does not affect IL-10 production. IL-12p70 and IL-10 concentration in the supernatants of human MDDC, treated as described in Table 1
, was measured by a specific sandwich ELISA. Data represent mean ± SEM of six independent experiments performed. *, P = 0.014; CTB-LPS-MDDC versus LPS-MDDC.
|
rCTB-treated MDDC show a reduced ability to stimulate MLR response
The results obtained with the analyses of the surface molecule expression and cytokine production induced us to test whether rCTB has any effect on MDDC immunostimulatory capacity.
Therefore, their ability to stimulate an allogeneic T cell response was assessed. To this purpose, graded numbers of CTB-LPS-MDDC and LPS-MDDC were cultured with a fixed number of allogeneic PBMC for 5 days. As expected, LPS-MDDC were quite efficient in inducing proliferation of alloreactive cells, and CTB-LPS-MDDC induced a significant (P<0.05 by Wilcoxon test, n=8), lower proliferative response (Fig. 2 ).
 View larger version (12K): [in a new window] |
Figure 2. rCTB pretreatment down-regulates the immunostimulatory capacity of LPS-MDDC in MLR. PBMC were cultured with graded numbers of allogeneic LPS-MDDC (•) or CTB-LPS-MDDC ( ). At Day 4, methyl 3H-thymidine (3H-TdR) was added for the last 18 h of culture. One representative experiment out of eight is shown.
|
 View larger version (13K): [in a new window] |
Figure 3. CD4+CD45RA+ T cells stimulated with CTB-LPS-MDDC show reduced proliferative capacity and increased production of IL-10 when stimulated with anti-CD3/anti-CD28 mAb. CD4+CD45RA+ T cells were cultured with allogeneic LPS-MDDC (TLPS-MDDC) or CTB-LPS-MDDC (TCTB-LPS-MDDC) at a 10:1 ratio. After 1 week, cells were collected, washed, stimulated with  CD3/CD28 mAb, and analyzed for proliferative capacity and cytokine production. (A) Proliferation of TLPS-MDDC (solid bar) and TCTB-LPS-MDDC (open bar). Columns represent mean ± SEM of three independent experiments. *, P = 0.01, TCTB-LPS-MDDC versus TLPS-MDDC. (B) Cytokine production by TLPS-MDDC (solid bars) and TCTB-LPS-MDDC (open bars) measured in supernatants of CD3/CD28-stimulated cells (24–48 h, see Materials and Methods). Columns represent the mean ± SEM of three independent experiments. *, P < 0.05, TCTB-LPS-MDDC versus TLPS-MDDC.
|
(inhibition: 57%; P< 0.05), and the same levels of TGF-β and IL-4 (Fig. 3B)
. Taken together, these results suggest that CTB-LPS-MDDC might induce the differentiation of Tregs.
CTB-treated MDDC induce functional CD4+ Tr1 cells
To evaluate the ability of TCTB-LPS-MDDC to inhibit autologous TLPS-MDDC proliferation, the latter cell population was restimulated with LPS-MDDC in the presence or absence of TCTB-LPS-MDDC (see Materials and Methods). TLPS-MDDC and TCTB-LPS-MDDC were labeled with CFSE and PKH-26, respectively. In preliminary experiments, different numbers of TCTB-LPS-MDDC were added to TLPS-MDDC cultures to identify the optimal ratio, which at 1:1, was chosen for the suppressive activity and for the IL-10 production evaluation (data not shown). As shown in Figure 4A
, TCTB-LPS-MDDC inhibited the proliferation of CFSE-labeled TLPS-MDDC consistently (P<0.05).
 View larger version (32K): [in a new window] |
Figure 4. TCTB-LPS-MDDC exert regulatory activity on TLPS-MDDC, which is cell–cell contact-independent. TLPS-MDDC and TCTB-LPS-MDDC, after 6 days of culture with rhIL-2, were labeled with CFSE or PKH-26, respectively. CFSE-labeled TLPS-MDDC were restimulated with LPS-MDDC at a 10:1 ratio in the presence or absence of TCTB-LPS-MDDC for 4 days. (A) Proliferation of CFSE-labeled TLPS-MDDC was assayed at Day 4 by flow cytometry and expressed as P.I., calculated by ModFit LT software. One representative experiment out of three is shown (P<0.05; TLPS-MDDC+LPS-MDDC vs. TLPS-MDDC+TCTB-LPS-MDDC+LPS-MDDC). (B) Parallel experiments were performed in a Transwell system that separated TLPS-MDDC and TCTB-LPS-MDDC. Transwell cultures with TLPS-MDDC in the upper and the lower chamber were used as control (see Materials and Methods). Proliferation of CFSE-labeled TLPS-MDDC was assayed at Day 4. One representative experiment out of three is shown (P<0.05; TLPS-MDDC+LPS-MDDC vs. TLPS-MDDC+TCTB-LPS-MDDC+LPS-MDDC).
|
As T cells stimulated with CTB-LPS-MDDC in the presence of anti-CD3/anti-CD28 mAb produced TGF-β and IL-10, we performed additional experiments in the presence of blocking mAb. As shown in Figure 5 , the addition of anti-IL-10R and anti-TGF-β mAb to the cultures completely reversed the inhibitory effect.
 View larger version (21K): [in a new window] |
Figure 5. TCTB-LPS-MDDC exert regulatory activity on TLPS-MDDC via IL-10 and TGF-β production. CFSE-labeled TLPS-MDDC were stimulated in the presence of PKH-26-labeled TCTB-LPS-MDDC as described in Figure 4
. TCTB-LPS-MDDC were tested for their ability to suppress the proliferation of CFSE-labeled TLPS-MDDC in the presence or absence of blocking IL-10R and TGF-β mAb. Proliferation of CFSE-labeled cells was assayed at Day 4 by flow cytometry and expressed as P.I. calculated by ModFit LT software. One representative experiment out of three is shown.
|
DISCUSSION
Many studies demonstrate that DC play an important role in the induction and maintenance of immune tolerance [20 , 21 ]. In this study, we found that pretreatment of human MDDC with rCTB reduces their ability to produce IL-12 upon challenge with LPS and affects their maturation with an impaired capability of alloantigen presentation. Furthermore, we analyzed the influence of CTB-LPS-MDDC on T cells. Our data demonstrate that CTB-LPS-MDDC induce the in vitro differentiation of IL-10-producing T cells with low proliferative capacity, which behave as Tr1 cells, described by different authors [13 14 15 , 22 ].
The CTB subunit, when conjugated with an antigen, is a highly efficient carrier molecule for the induction of mucosal antibody responses [23 , 24 ] and oral tolerance [25 ]. The mechanism underlying this effect is probably a result of the ability of the CTB to promote the presentation of the coupled antigen with a concomitant increase of CD40 and CD86 expression on APC [25 ]. The apparently opposite outcome produced by the oral administration of CTB-antigen conjugates might be reconciled by the notion that systemic tolerance in vivo induced by feeding with CTB-conjugated antigens may be associated with a mucosal antibody response [26 , 27 ]. However, the observation that systemic tolerance is associated with production of regulatory cytokines (i.e., TGF-β) suggests that CTB might have intrinsic, immunomodulatory properties [5 ].
Indeed, it has been demonstrated that CTB, even in an unconjugated form, is able to increase the specific tolerogenic effect of oral insulin in a murine model of autoimmune diabetes [28
]. Moreover, it has been shown that orally administered rCTB prevents development of TNBS-induced colitis in mice and brings about resolution of previously established colitis by inhibiting mucosal IL-12 and IFN- production, suggesting an immunomodulatory role of free CTB on DC [10
].
In the present study, we demonstrate for the first time that CTB is able to influence the phenotype and functions of human DC. Specifically, we observed that preincubation of immature MDDC with rCTB partially prevents IL-12 production and the full expression of CD40 after LPS stimulation, without affecting the expression of costimulatory molecules CD80, CD86, and CD83. These results are in agreement with a previous observation made in the murine system, whereby free CTB, differently from CTB conjugated with antigens, was found not to enhance the cell-surface levels of CD80 and CD86 [7 ]. However, they are in contrast with other data reporting the effect of CTB on splenic murine DC, which demonstrated that the in vitro incubation of murine DC with rCTB induces up-regulation of the expression of MHC class II and CD86 on DC and an increased secretion of IL-12 [29 ]. Reasons for these discrepancies are not easily recognized, but heterogeneity of myeloid murine DC and their state of maturation might account for the contrasting results observed in the murine system [30 , 31 ]. We observed that as opposite as in the case of CTB-antigen conjugates [25 ], free CTB exerts a slight but statistically significant effect on the CD40 expression on MDDC. However, recently, it has been demonstrated that murine DC lacking CD40 expression induce the differentiation of CD4+ Tregs that are capable of producing IL-10 in an antigen-specific manner, thus suggesting that CD40 could be a key determinant of the decision between tolerance and immunity [32 ].
Taken together, the data suggest that CTB is able to interfere with the maturation-activation process of DC, inducing a cell quite similar to the DC treated with different biological and pharmacological agents described to be able to induce Tregs [33 34 35 ]. As reviewed recently, tolerogenic DC show no changes of the expression of canonical costimulatory molecules and no changes of IL-10 production, whereas IL-12 production remains low upon LPS stimulation [14 ]. Similar effects were recently observed in a human-derived macrophage cell line and human PBMC incubated with CTB [8 ]. Previous data have been described in the murine system, where CT-treated bone marrow-derived DC have the ability to induce Tregs [36 , 37 ]. In consideration of our results, we assume that this effect may be ascribed prevalently to the B subunit of the toxin.
In the present study, MDDC treated with rCTB are indeed able to generate Tregs from CD4+CD45RA+ T cells in vitro. These T cells, when compared with T cells treated with LPS-MDDC, showed reduced ability to proliferate, reduced IFN- production, increased IL-10 production, and showed the same levels of TGF-β upon polyclonal stimulation. Furthermore, TCTB-LPS-MDDC do not express high levels of Foxp3 and exert their ability to suppress the specific TLPS-MDDC proliferation through the immunosuppressive cytokines IL-10 and TGF-β. These properties are similar to those described for Tr1 cells.
Groux et al. [22 ] first described the biological features of Tr1. In the past few years, different populations of IL-10-producing T cells with regulatory function have been described [38 39 40 41 42 43 44 ]. Their cytokine profile can slightly vary depending on the experimental conditions, but their levels of IL-10 production are invariably high. It has also been established that Foxp3 is a marker for natural CD4+CD25+ but not for adaptive Tr1 cells [14 ].
Several observations indicate that Tr1 cells are involved in maintaining peripheral tolerance in vivo. Tr1 cells play a critical role in modulating immune responses to self-antigens. Reduction of IL-10-producing CD4+ T cells was observed in peripheral blood and sinovial tissue of patients with rheumatoid arthritis [45 ] and peripheral blood of patients with autoimmune hemolytic anemia [46 ].
Tr1 cells can also regulate immune responses to allergen and mucosal antigens. In particular, the ability of Tr1 cells to down-modulate responses to nickel [47 ], insect venom [48 ], and cat allergen [49 ] has been reported. Furthermore, Tr1 cells have been generated in a murine system from lamina propria T cells after incubation with cecal bacterial antigen [50 ]. Tr1 cells also have the ability to prevent the development of colitis when cotransferred with pathogenic CD4+CD45RBhi T cells in the SCID transfer model of colitis [22 ]. In humans, Tr1 clones have been generated recently from lymphocytes isolated from intestinal mucosa of patients with celiac disease in clinical remission [51 ]. Furthermore, spontaneous development of tolerance in severe combined immunodeficient patients transplanted with HLA-mismatched allogeneic stem cells [52 53 54 55 56 ] and in kidney or liver transplantation [57 ] is associated to high levels of IL-10 produced by T cells, indicating that Tr1 cells can induce tolerance in bone marrow and solid organ transplantation.
Therefore, many strategies have been adopted to induce/expand these cells to control immune responses in different pathologies. Immunological tolerance can be established by selected immunomodulatory compounds through the induction of Tr1 cells. Administration of rapamycin and IL-10 [15 ] or deoxyspergualin and anti-CD3 immunotoxin, which associates with elevated levels of serum IL-10 [58 ], induces an antigen-specific, long-term tolerance as a result of down-regulation of inflammation, blockade of effector T cells, and generation of Tr1 cells in diabetic animals successfully transplanted with pancreatic islets.
In conclusion, this study reports the first evidence in humans of the CTB ability to generate Tr1 cells in vitro through affecting DC maturation induced by LPS. Whether this effect on MDDC is restricted to TLR4 stimulation or might involve other TLRs was not investigated in the present study. However, our preliminary results suggest that MDDC stimulation through TLR2 might exert similar effects. Further studies are necessary to clarify this issue.
Administration of Tregs, generated in vitro by MDDC pretreated with CTB in a culture system suitable for human use, might represent a valuable mean for the modulation of immune responses in T cell-mediated diseases.
ACKNOWLEDGEMENTS
This work was supported by Ministero della Salute, Grant 6AFC, 6AFC/2, Italy, to M. B. The authors have no financial conflict of interest. The authors thank Doriana Campanile, Cinzia Butteroni, and Corrado Volpe for excellent technical assistance, Dr. Angelo De Milito for his support for ModFit software analysis, and Dr. Bianca Barletta for helpful discussion.
FOOTNOTES
1 These authors contributed equally to this work. 
2 These authors share the senior authorship of this paper.
Received December 21, 2007; revised April 14, 2008; accepted May 23, 2008.
REFERENCES
} production and signaling in experimental colitis and Crohn’s disease Gut 54,1558-1564 J. Exp. Med. 179,1109-1118
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
