Constitutive Expression of IDO by Dendritic Cells of Mesenteric Lymph Nodes: Functional Involvement of the CTLA-4/B7 and CCL22/CCR4 Interactions

Toshiharu Onodera, Myoung Ho Jang, Zijin Guo, Mikako Yamasaki, Takako Hirata, Zhongbin Bai, Noriko M. Tsuji, Daisuke Nagakubo, Osamu Yoshie, Shimon Sakaguchi, Osamu Takikawa and Masayuki Miyasaka
J Immunol November 1, 2009, 183 (9) 5608-5614; DOI: https://doi.org/10.4049/jimmunol.0804116

Abstract

Dendritic cells (DCs) express the immunoregulatory enzyme IDO in response to certain inflammatory stimuli, but it is unclear whether DCs express this enzyme under steady-state conditions in vivo. In this study, we report that the DCs in mesenteric lymph nodes (MLNs) constitutively express functional IDO, which metabolizes tryptophan to kynurenine. In line with a previous report that regulatory T cells (Tregs) can induce IDO in DCs via the CTLA-4/B7 interaction, a substantial proportion of the MLN DCs were located in juxtaposition to Tregs, whereas this tendency was not observed for splenic DCs, which do not express IDO constitutively. When CTLA-4 was selectively deleted in Tregs, the frequency of IDO-expressing DCs in MLNs decreased significantly, confirming CTLA-4’s role in IDO expression by MLN DCs. We also found that the MLN DCs produced CCL22, which can attract Tregs via CCR4, and that the phagocytosis of autologous apoptotic cells induced CCL22 expression in CCL22 mRNA-negative DCs. Mice genetically deficient in the receptor for CCL22, CCR4, showed markedly reduced IDO expression in MLN-DCs, supporting the involvement of the CCL22/CCR4 axis in IDO induction. Together with our previous observation that MLN DCs contain much intracytoplasmic cellular debris in vivo, these results indicate that reciprocal interactions between the DCs and Tregs via both B7/CTLA-4 and CCL22/CCR4 lead to IDO induction in MLN DCs, which may be initiated and/or augmented by the phagocytosis of autologous apoptotic cells by intestinal DCs. Such a mechanism may help induce the specific milieu in MLNs that is required for the induction of oral tolerance.

Because the intestine is continually challenged by a wide variety of Ags, the intestinal APCs have to control the most complex situations by promoting or suppressing T cell responses, depending on the circumstances (1). Not only do they have to distinguish invasive organisms from harmless Ags, but they must also regulate the extent of the immune responses. In this regard, the dendritic cells (DCs)3 in mesenteric lymph nodes (MLNs) have been implicated in the regulation of immune responses against commensal bacteria and food Ags and are apparently essential for the generation of local and systemic tolerance to these Ags, which is known as oral tolerance (2, 3).

A number of mechanisms have been suggested for the role of DCs in tolerance induction. For instance, immature DCs lacking sufficient costimulatory molecule expression induce regulatory T cells (Tregs), whereas mature DCs expressing high levels of costimulatory molecules induce immune activation (4). DCs can also control tolerance induction by their ability to produce anti-inflammatory cytokines, such as IL-10 (5) and TGF-β (6). In addition, certain differentiated DCs in the intestinal lamina propria (LP) have been implicated in suppressing T cell activation by inducing CD4+ T cells to produce IL-10 (5, 7, 8). Recent studies indicate that DCs also secrete retinoic acid to enhance the TGF-β-dependent conversion of naive T cells into Tregs (9, 10, 11) and to induce Th17 cells (12).

Tolerogenic DCs also employ IDO, a rate-limiting enzyme that initiates the catabolism of tryptophan (13). IDO-competent DCs exert tolerogenic effects on T cells that are mediated by tryptophan depletion (14) and by the production of kynurenines (15). IDO has been implicated in immunological tolerance (16) and in immunosuppression in pregnancy (17), tumor resistance (18), chronic infection (19), and autoimmunity (20). To acquire IDO activity, DCs require a specific gene expression profile and appropriate stimuli, such as IFN-γ and/or soluble recombinant CTLA-4 (CTLA-4-Ig) (21, 22). CTLA-4 activity is important for Treg-induced tolerance in several settings (23, 24). Although the constitutive CTLA-4 expression by Tregs has been shown to mediate an IDO-inducing effect in vitro (25), it remains to be fully elucidated whether and under which circumstances this mechanism occurs in vivo.

In this study, we report that the active form of IDO is constitutively expressed by MLN DCs. In line with a previous report that Tregs can induce IDO in DCs via an interaction between CTLA-4 of Tregs and B7 on DCs (25), MLN DCs were frequently located in juxtaposition to Tregs. The selective deletion of CTLA-4 in Tregs significantly decreased the frequency of IDO-expressing DCs in MLNs. MLN DCs also produce CCL22, which can attract Tregs (26), and we found that the phagocytosis of autologous apoptotic cells induced CCL22 expression in CCL22 mRNA-negative DCs. Mice deficient in the receptor for CCL22, CCR4, showed markedly reduced numbers of IDO+ DCs in MLNs, suggesting the functional involvement of the CCL22/CCR4 axis in IDO induction. Collectively, these results indicate that intricate interactions between the DCs and Tregs via both B7/CTLA-4 and CCL22/CCR4 lead to the induction of IDO in MLN DCs and that these interactions may be induced, at least in part, by the uptake of autologous apoptotic cells by the DCs. Such a mechanism may help generate the unique tolerogenic milieu within MLNs.

Materials and Methods

Mice

Female BALB/c mice were obtained from CLEA Japan. Ctla-4flox/flox × Foxp3IRES-Cre mice on the BALB/c background (27) were maintained in the animal facilities of the Kyoto University Graduate School of Medicine and used at 6 wk of age. The spleen and lymph nodes of these mice were enlarged, but the general architecture was well maintained, and the localization of DCs, Tregs, and B cells was very similar to those of the littermate control and wild-type mice. Ccr4−/− mice were obtained from The Jackson Laboratory, backcrossed to the C57BL/6 background, and maintained in the animal facilities of the Kinki University School of Medicine. The mice were maintained under specific pathogen-free conditions. All animal experiments were performed under experimental protocols approved by the Ethics Review Committee for Animal Experimentation of the Osaka University Graduate School of Medicine.

Preparation of DCs from MLNs, small intestinal LP, Peyer’s patches (PPs), and spleen

DCs were prepared as described previously (7). In brief, segments of small intestine and PPs were treated with PBS containing 10% FCS, 20 mM HEPES, 100 U/ml penicillin, 100 μg/ml streptomycin, 1 mM sodium pyruvate, 10 mM EDTA, and 10 μg/ml polymyxin B (Calbiochem) for 30 min at 37°C to remove epithelial cells and were washed extensively with PBS. The small intestinal segments, PPs, MLNs, and spleen were digested with collagenase D (Roche) and DNase I (Roche) with continuous stirring at 37°C for 45 min. EDTA was added, and the cell suspension was incubated for an additional 5 min at 37°C. Cells were spun through 15.5% Accudenz (Accurate Chemical and Scientific) solution to obtain a population rich in DCs. The obtained cells were incubated with FITC-conjugated anti-I-Ad or FITC-conjugated anti-B220 and PE-conjugated anti-CD11c after FcR blocking. DC subsets were sorted on the basis of their expression of CD11c and I-Ad by a FACSVantage SE (BD Biosciences). The purity of the sorted DCs was >95%.

In vitro uptake of apoptotic epithelial cells by DCs

Small intestinal epithelial cells (IECs) were obtained as described previously (28). The IECs were cultured for 2 h to induce spontaneous apoptosis. Cells obtained from the low-density Accudenz fraction (containing DCs) were mixed with apoptotic IECs and cultured for 4 h at 37°C. The DCs were then isolated from the mixture by cell sorting on a FACSVantage SE (BD Biosciences) and cultured for 40 h in the presence or absence of CCL19 (R&D Systems). The amount of CCL22 in the supernatant was then determined by ELISA according to the manufacturer’s instructions (R&D Systems).

Quantitative PCR for IDO and CCL22 expression

The total RNA was prepared from MLN DCs, LP DCs, PP DCs, or splenic (SP) DCs using TRIzol reagent (Invitrogen/Life Technologies). The RNA was then reversed-transcribed using Superscript II (Invitrogen) according to the manufacturer’s instructions, with random hexamers as primers. Quantitative PCR analysis was performed using the 7700 Sequence Detector (Applied Biosystems). The Quantitative PCR was conducted in a final volume of 25 μl containing cDNA, 2× PCR Master Mix (Applied Biosystems), and primers specific for IDO (Mm00492586_m1; Applied Biosystems) or CCL22 (Mm00436439_m1; Applied Biosystems). β-Actin was measured as an internal control. The amplification conditions were 95°C for 15 s, 60°C for 60 s, and 50°C for 120 s. As a positive control for IDO expression, DCs obtained from the spleen were exposed overnight to 200 U/ml rIFN-γ (R&D Systems) and 20 μg/ml CTLA-4-Ig (BD Pharmingen).

Immunohistochemistry

Tissue sections were fixed in methanol at −30°C for 10 min, then treated with a Biotin Blocking System (DakoCytomation), followed by incubation with 5 μg/ml purified rat anti-mouse CD16/CD32 mAb (BD Pharmingen) for 1 h at room temperature. To detect IDO+ DCs, rabbit anti-mouse IDO polyclonal Abs (pAbs) (two different pAbs were used: one was generated by one of us, O.T., and the other was a gift from Dr. A. L. Mellor, Medical College of Georgia Augusta, GA; both Abs yielded basically the same results). To detect CD11c, we used a biotin-conjugated hamster anti-mouse CD11c mAb (BD Pharmingen). To determine the localization of Tregs, a rabbit anti-mouse Foxp3 pAb (a gift from Dr. K. Matsushima, University of Tokyo, Tokyo, Japan) was used instead of the rabbit anti-mouse IDO Ab as the primary Ab. Specificity control for the IDO pAb included using preimmune rabbit serum as the primary Ab, whereas for Foxp3, rabbit IgG (Chemicon International) was used. The primary Abs were detected with Alexa Fluor 488-labeled chicken anti-rabbit IgG (Molecular Probes) and streptavidin-Alexa Fluor 594 conjugate (Molecular Probes). The nuclei were then stained with Hoechst 33258 (Invitrogen).

Measurement of IDO enzymatic activity

The IDO enzyme assay was performed as previously reported (29), with some modifications. In brief, freshly isolated DCs were washed, resuspended in sterile HBSS (Sigma-Aldrich) containing 500 μM tryptophan (Sigma-Aldrich), and incubated for 4 h. The supernatants were then harvested and assayed for kynurenine. For the assay, 30 μl of 30% trichloroacetic acid was added to 60 μl of culture supernatant and the mixture was vortexed and centrifuged at 10,000 × g (12,000 rpm) for 5 min. Then, 40 μl of supernatant was added into an equal volume of Ehrlich reagent (5 μl of glacial acetic acid and p-dimethylaminobenzaldehyde). The OD was measured at 492 nm using a NanoDrop (LMS). Purified l-kynurenine (0–500 μM; Sigma-Aldrich) was used as a standard.

Statistical analysis

Data are presented as the mean ± SD as indicated. The cells were isolated from more than five mice in each experiment. When necessary, a two-group comparison was performed using Student’s t test. Values of p < 0.05 were considered statistically significant.

Results

MLN DCs express IDO under physiological conditions

Although the small intestine is continuously exposed to gut flora and dietary Ags, the intestinal immune system almost invariably succeeds in balancing protective immune responses and tolerance. To examine whether DCs utilize IDO as one mechanism for inducing mucosal tolerance, we performed immunohistochemical staining for IDO and CD11c on tissue sections from the MLNs and spleen of experimentally naive mice. As shown in Fig. 1⇓A, under physiological conditions, IDO staining (green) was clearly observed in a substantial proportion of the CD11c+ DCs (red) in the MLNs, but not in the spleen, in agreement with a previous study (30). These DCs were CD11chigh and were not plasmacytoid DCs, as described later. To quantify the expression level of IDO in different populations of DCs, we purified the MLN DCs and SP DCs by cell sorting on the basis of their expression of CD11c and MHC class II (Fig. 1⇓B) and performed quantitative PCR analyses. As shown in Fig. 1⇓C, IDO transcripts were abundantly expressed in the CD11chighMHC class IIhigh MLN DCs at >100 times the level in splenic DCs, confirming the constitutive and selective expression of IDO by MLN DCs. We next examined whether the DC-expressed IDO was functional. As shown in Fig. 1⇓D, the culture supernatant from MLN DCs contained high levels of kynurenine, whereas the culture supernatant obtained from SP DCs in the same way showed only a background level of this molecule, demonstrating that the IDO expressed by MLN DCs was functional.

FIGURE 1.
FIGURE 1.

MLN DCs contain an IDO-positive subpopulation under physiological conditions. Frozen tissue sections were prepared and immunostained as described in Materials and Methods. Briefly, the MLN and spleen were stained with an anti-IDO pAb and anti-CD11c mAb. The stained cells were labeled by immunofluorescence to show IDO protein (green) in DCs (red). The nuclei were then stained with Hoechst 33258 (blue). MLN DCs included a subpopulation that constitutively synthesized IDO (A, left panel). IDO+ cells were absent from the spleen (A, right panel). Purity of the low-density fraction MLN DCs stained for CD11c and MHC class II before and after sorting (B). The expression of IDO in MLN DCs and SP DCs was evaluated by real-time PCR. cDNA was prepared from the total RNA obtained from freshly FACS-sorted MLN DCs and SP DCs. IDO was expressed in the MLN DCs but not in the SP DCs (C). The IDO activity in freshly isolated DCs was measured as kynurenine production in the culture supernatant. CD11chighMHC class IIhigh DCs from the MLN and spleen were sorted and incubated in HBSS for 4 h (D). Studies were performed in triplicate and repeated twice with similar results. A representative experiment is shown. Error bars indicate SD.

IDO expression in MLN DCs is dependent on some factors unique to the MLN microenvironment

Juxtaposition of Tregs and DCs.

As one mechanism by which MLN DCs express high levels of IDO, we speculated that IDO expression is induced via a cognate interaction with Tregs. The constitutive CTLA-4 expression by Tregs has been shown to induce IDO activity in DCs through the ligation of B7 molecules in vitro (31). Moreover, in the mucosal immune system, MLN DCs and MLN Tregs showed a high expression of B7 and CTLA-4, respectively (data not shown). We therefore investigated whether MLN DCs are physically associated with Tregs in MLNs. As shown in Fig. 2⇓A, the outer cortex of the MLNs contained a diffuse population of Tregs and DCs, which were frequently juxtaposed. In contrast, most of the DCs in the spleen were separated from the Tregs: the SP DCs were mainly located in the marginal zone and scarce in the T cell zone, whereas the Tregs resided principally within the T cell zone of the white pulp. Peripheral lymph nodes (PLNs) also showed the sparse population of DCs in contrast to the relatively dense population of Tregs. We next assessed the proportion of DCs that were located in close proximity to Tregs in the MLNs, spleen, and PLNs and found that >60% of the DCs were associated with Tregs in the MLNs, whereas <20% of the DCs exhibited this association in the spleen. PLN DCs made up a much smaller population in comparison to MLN DCs and about half of them were in close proximity to Tregs (Fig. 2⇓B). To further investigate the distribution of IDO+CD11c+ DCs and Tregs, we examined serial sections of MLNs stained for IDO, CD11c, and Foxp3 and found that a majority of IDO+ DCs were in juxtaposition to Tregs in MLNs (Fig. 2⇓C).

FIGURE 2.
FIGURE 2.

MLN DCs express IDO in a manner dependent on CTLA-4 on Tregs and interact with Tregs preferentially. Frozen tissue sections of the spleen and MLN were fixed and stained with an anti-Foxp3 pAb and biotinylated anti-CD11c mAb. The sections were further incubated with fluorescently labeled reagents to reveal the distribution of Foxp3+ Tregs (green) and DCs (red). In MLNs (A, upper panels), the Tregs were distributed widely in the parafollicular cortex and were often in close contact with MLN DCs. On the other hand, in the spleen (A, middle panels), the Tregs were frequently seen in the T cell zone, while the DCs were localized chiefly along the marginal zone, and they showed little association geographically. In PLNs, Tregs were more widely distributed and a small proportion of them were in close proximity to DCs compared with MLNs (A, lower panels). B, The total number of DCs and the number of DCs showing close contact with Tregs per unit area (mm2) were determined in the above sections, and the data are shown as a bar graph. Error bars indicate SD. C, MLN serial sections were stained for IDO, CD11c, and Foxp3; a majority of IDO+ DCs were in juxtaposition to Tregs in MLNs. D, The frequency of IDO+CD11c+ cells among CD11c+ MLN DCs was compared between wild-type mice and Ctla-4flox/flox × Foxp3IRES-Cre mice, which show a selective deficiency of CTLA-4 in Tregs (left). ∗∗, p < 0.0005 (left panel). The IDO activity was assessed by kynurenine production in the culture supernatant of DCs derived from Ctla-4flox/flox × Foxp3IRES-Cre mice (Ctla-4flox/flox) and Ctla-4flox/wt × Foxp3IRES-Cre (Ctla-4flox/wt) mice. CD11chighMHC class IIhigh DCs were sorted from MLNs and spleen by FACS and incubated in HBSS for 4 h (right panel). This study was performed in triplicate. Statistics were performed using a one-tailed Student’s t test. ∗, p < 0.05. Error bars indicate SD.

Involvement of CTLA-4.

We next assessed the contribution of CTLA-4 to the IDO expression by MLN DCs, because CTLA-4 expression by Tregs has been shown to mediate an IDO-inducing effect in vitro (25, 31). As shown in Fig. 2⇑D (left), Ctla-4flox/flox × Foxp3IRES-Cre mice, whose Tregs are selectively deficient in CTLA-4 expression (27), exhibited an ∼50% reduction in IDO+CD11c+ DCs compared with the level in wild-type control mice, although the extent of the geographical association between Tregs and DCs in the MLNs was comparable to that seen in wild-type mice (data not shown). When IDO activity was assessed by the levels of kynurenine in the culture supernatant of MLN DCs, those from Ctla-4flox/flox × Foxp3IRES-Cre mice showed significantly lower IDO activity than those from littermate Ctla-4flox/wt × Foxp3IRES-Cre control mice (Fig. 2⇑D, right). These findings indicate that CTLA-4 plays an important role in IDO induction in MLN DCs but is dispensable for the juxtapositioning of MLN Tregs with MLN DCs.

Involvement of the CCL22/CCR4 interaction.

We previously reported that the small intestinal LP DCs produce CCL22 and might be involved in the recruitment of Tregs to the LP (32). We therefore thought it possible that MLN DCs also employ a similar, if not identical, strategy to recruit Tregs into their vicinity. In line with this hypothesis, MLN DCs expressed a high level of CCL22 mRNA, which was comparable to that observed in LP DCs, whereas the DCs from PPs and spleen expressed very little CCL22 mRNA (Fig. 3⇓A). Furthermore, the CCL22 produced by the MLN DCs was functionally active, because the culture supernatant of MLN DCs attracted Tregs, and this was specifically blocked by an anti-CCL22 mAb (data not shown). These results point to the possibility that MLN DCs utilize CCL22 to interact with CCR4-expressing Tregs in MLNs, which may also contribute to the induction of IDO in DCs. To address this possibility, we asked whether CCL22 and IDO are expressed in the same MLN DC subset. Our previous work indicated that MLN DCs consist of at least four phenotypically distinguishable subsets (Fig. 3⇓B, a, CD11chighCD8αhighβ7int; b, CD11chighCD8αintβ7high; c, CD11chighCD8α−β7high; and d, CD11chighCD8α−β7low), two of which (Fig. 3⇓B, b and c) are derived from the small intestinal LP and the rest are from PPs (7). We therefore purified these four DC subsets and performed a quantitative real-time PCR analysis. As shown in Fig. 3⇓B, the LP-derived β7 integrinhigh MLN DC subset (c) showed elevated expression of both CCL22 and IDO at the mRNA level, in agreement with the hypothesis that CCL22 and IDO are coexpressed in a DC subset in MLNs. We next asked whether the CCL22 expression is functionally related to the MLN DC IDO expression and to their interaction with Tregs. To this end, we used CCR4-deficient mice. As shown in Fig. 3⇓C (left), Ccr4−/− mice showed a markedly reduced frequency of IDO+CD11c+ DCs, although the overall pattern of association between Tregs and DCs in the MLNs was comparable to that in wild-type mice (data not shown). IDO activity of Ccr4−/− mice was found to be about half of Ccr4+/− MLN DCs, as judged by the kynurenine concentration in the DC culture supernatant (Fig. 3⇓C, right). These results are compatible with the hypothesis that CCR4 signaling in Tregs plays an important role in the induction of IDO in MLN DCs, but show that it is dispensable for the juxtapositioning of MLN Tregs with MLN DCs.

FIGURE 3.
FIGURE 3.

Expression of CCL22 and IDO in DC subsets. A, cDNA was prepared from the total RNA obtained from freshly FACS-sorted MLN DCs, LP DCs, PP DCs, and SP DCs. The expression of CCL22 was quantified by real-time PCR. Samples were normalized individually to the LP DCs. Studies were performed in triplicate and repeated twice with similar results. A representative experiment is shown. ∗∗∗, p < 0.0001 for MLN DCs vs LP DCs. Error bars indicate SD. B, Four DC subsets in the MLNs (a, CD11chighCD8αhighβ7int; b, CD11chighCD8αintβ7high; c, CD11chighCD8α−β7high; and d, CD11chighCD8α−β7low) were FACS sorted based on their CD11c, CD8α, and β7 integrin expression, and the expression of IDO and CCL22 was analyzed by real-time PCR. In C, the frequency of IDO+CD11c+ cells among CD11c+ MLN DCs was compared between wild-type mice and Ccr4−/− mice (left panel). ∗∗∗, p < 0.0005. The IDO activity was assessed by kynurenine production in the culture supernatant of DCs derived from Ccr4+/− mice and Ccr4−/− mice (right panel). This study was performed in triplicate. ∗∗, p < 0.005. Error bars indicate SD.

CCL22 is induced in DCs upon the uptake of apoptotic cells and chemokine stimulation

To gain insight into the mechanism by which a MLN DC subset coexpresses high levels of CCL22 and IDO, we focused on the specific environment in the intestine, where self -Ags from spontaneously apoptotic epithelial cells are likely to be abundant along with other types of Ags. Given that the uptake of apoptotic cells up-regulates CCR7 in DCs (33) and that CCR7 signaling is critical to the DC migration from the LP into the MLNs (7), we examined whether the engulfment of apoptotic cells and signaling through CCR7 induced CCL22 and IDO in DCs. For this purpose, we used SP DCs, because they do not produce CCL22 or IDO at significant levels under physiological conditions, as described above. As shown in Fig. 4⇓, either the addition of apoptotic cells or stimulation with CCL19 alone substantially up-regulated the CCL22 production by the DCs. Furthermore, the stimulation of DCs with both apoptotic cells and CCL19 markedly potentiated the production of CCL22 by the DCs, whereas IDO was not detectable in the DCs under the experimental conditions we used (data not shown). Although the apparent absence of IDO induction requires further verification, these results indicate that CCL22 expression can be triggered by the uptake of apoptotic cells by DCs and that IDO expression may be induced by a different mechanism.

FIGURE 4.
FIGURE 4.

Induction of CCL22 in DCs by exposure to apoptotic cells and CCR7 ligand. SP DCs and plasmacytoid DCs were cocultured in the presence or absence of apoptotic IECs for 4 h. After sorting, the DCs were cultured with or without CCL19 overnight. Total RNA was subsequently prepared from these DCs and the CCL22 mRNA was quantified by quantitative PCR analysis as described in Materials and Methods. The results shown here are the means ± SD of two different experiments, each performed in triplicate. ∗∗∗, p < 0.0001. Error bars indicate SD.

Discussion

Accumulating evidence indicates that IDO plays an important role in maternal tolerance to the allogeneic fetus (17), self-tolerance in nonobese diabetic mice (34), and, in certain pathological conditions, including inflammatory bowel disease (35). IDO is inducible in many cell types and, in particular, its expression in DCs has been considered to be crucial in the suppression of T cell proliferation and survival (14). However, the regulatory DC subset that induces tolerance in an IDO-dependent manner has not been specified. In the present study, we showed that MLN DCs bearing the CD11chighCD8α−β7high phenotype constitutively expressed the active form of IDO and that the B7/CTLA-4 and CCL22/CCR4 interactions contributed to the IDO expression in the MLN DCs. These DCs were clearly different from plasmacytoid DCs, because they were CD11chigh and B220−.

Although previous studies have indicated the involvement of CTLA-4 expressed by Tregs in IDO induction in DCs in vitro (25), it was unclear whether this is also the case in vivo, since CTLA-4-deficient mice or the systemic injection of CTLA4-Ig only cause a net effect of the depletion/dysfunction of CTLA-4 on multiple cell types. Therefore, in the present study, we used Ctla-4flox/flox × Foxp3IRES-Cre mice, which are selectively deficient in CTLA-4 expression in Tregs, and found that the IDO expression in the MLN DCs was substantially decreased compared with the wild-type control MLN DCs (Fig. 2⇑D, right), although the frequency of the Tregs appeared uncompromised in the MLNs of the mutant mice. In addition, when compared with littermate control (Ctla-4flox/wt × Foxp3IRES-Cre), Ctla-4flox/flox × Foxp3IRES-Cre mice showed significantly lower IDO activity in MLN DCs. The Ctla-4flox/flox × Foxp3IRES-Cre mice showed enlarged spleen and lymph nodes with increased numbers of IFN-γ+ cells (27) that have the ability to induce IDO; this might have at least in part reduced the effect of CTLA4 deficiency on IDO activity in MLN DCs. These data collectively provide in vivo confirmation of the importance of Treg CTLA-4 in IDO induction in MLN DCs and indicate that additional molecular interactions, i.e., other than those mediated by CTLA-4/B7, are required for full IDO expression by the MLN DCs.

In addition to the CTLA-4/B7 interaction, the CCL22/CCR4 interaction appears to contribute to the IDO expression in MLN DCs, because CCL22 was expressed in a MLN DC subset that also expressed IDO, and the IDO-expressing MLN DCs were markedly reduced in Ccr4−/− mice. The MLN DCs expressing both CCL22 and IDO are judged to be derived from the small intestinal LP, because 1) LP DCs produce CCL22 (32), 2) LP DCs migrate to MLNs (7), and 3) MLN DCs with the LP DC phenotype (CD11chighCD8α−β7high) expressed both CCL22 and IDO, as shown in the present study. With regard to the constitutive production of CCL22 by MLN DCs, we speculate that the ingestion of apoptotic cells may play an important role. We previously reported that MLN DCs with the LP DC phenotype are poorly phagocytic but show irrefutable evidence of previous phagocytosis, containing much cellular debris within the cytoplasm in vivo (7), in agreement with this hypothesis. In the present study, we showed experimentally that CCL22 could be induced in DCs after the engulfment of apoptotic cells and that the CCL22 induction was further enhanced by the stimulation of the DCs with a DC maturation-promoting chemokine, CCL19. Given that the uptake of apoptotic cells up-regulates CCR7 in DCs (33) and that CCR7 signaling is important for the migration of LP DCs into MLNs (7), the ingestion of apoptotic cells in the presence of CCR7 ligands in the small LP may cause a variety of events, leading to the induction of CCL22 in LP DCs and their subsequent migration into MLNs. On the other hand, IDO induction may well be regulated by a different mechanism, since IDO is expressed only weakly, if at all, by LP DCs but is clearly expressed by MLN DCs. In addition, although the data are preliminary, we did not observe IDO induction in DCs after the ingestion of apoptotic cells under our experimental conditions. Although this observation appears to be in contrast to the study by Williams et al. (36), who reported the induction of IDO subsequent to the ingestion of apoptotic cells in human monocyte-derived DCs, they were able to induce IDO only in the presence of LPS stimulation. In our study, the DCs were not stimulated with LPS. Whether IDO can be induced in DCs that ingested apoptotic cells in the presence of LPS or other environmentally associated molecules remains to be examined.

Finally, a provocative hypothetical model of the sequence of events involving the induction of IDO and CCL22 in MLN DCs can be constructed from the results of the present study, as follows (Fig. 5⇓). 1) The ingestion of apoptotic epithelial cells derived from the normal turnover of tissues up-regulates CCR7 in LP DCs and the engulfment of apoptotic cells in conjunction with the enhanced CCR7 signaling up-regulates CCL22 in these LP DCs. 2) The stimulated LP DCs then enter the mesenteric lymphatics in a CCR7-dependent manner (the lymphatic endothelium has been reported to produce CCR7 ligands (37)) and then migrate into the MLNs where they form a CCL22-producing DC subset. 3) In the MLNs, the CCL22-positive MLN DCs act on Tregs via CCR4, which promotes the CTLA-4/B7 interaction between Tregs and DCs in situ. 4) This interaction finally induces IDO expression in DCs, possibly via the action of IFN-γ, as reported by others (21, 25). Although much work is needed to test this hypothesis, our study indicates that IDO is induced via intricate molecular interactions, including those mediated by CTLA-4/B7 and CCL22/CCR4 in MLN DCs. CCR7 signaling may also be involved in this process via the enhanced production of CCL22 in LP and MLN DCs. We hope that the present findings will lay the groundwork for further studies that will elucidate the precise role of IDO in oral tolerance in which MLNs play an essential role (38).

FIGURE 5.
FIGURE 5.

Proposed model for constitutive IDO expression in MLN DCs. A, Schematic representation of cellular events involved in the production of CCL22 and IDO in the small intestine/MLNs. B, The panels illustrate enlarged views of potential events occurring in each compartment depicted in A. See text for details.

Acknowledgments

We thank Drs. Satoshi Ueha and Kouji Matsushima (University of Tokyo, Tokyo, Japan) for providing the rabbit anti-mouse Foxp3 pAb. We also thank Dr. Andrew L. Mellor (Medical College of Georgia) for providing rabbit anti-mouse IDO pAbs. We thank members of the DNA Chip Development Center for Infectious Diseases (RIMD, Osaka University, Osaka, Japan) for technical advice. We also thank Shinobu Yamashita and Miyuki Komine for secretarial assistance.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 This work was supported by a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan (18790338 and 19041044 to M.H.J.) and a grant for Advanced Research on Cancer from the Ministry of Education, Culture, Sports, Science and Technology of Japan (1701456 to M.M.).

  • ↵2 Address correspondence and reprint requests to Dr. Myoung Ho Jang, Laboratory of Gastrointestinal Immunology, WPI Immunology Frontier Research Center, Osaka University, 3-1, Yamada-oka, Suita 565-0871, Japan and Dr. Masayuki Miyasaka, Laboratory of Immunodynamics, Department of Microbiology and Immunology, Graduate School of Medicine, Osaka University, C8, 2-2, Yamada-oka, Suita 565-0871, Japan. E-mail addresses: jang{at}frec.osaka-u.ac.jp and mmiyasak{at}orgctl.med.osaka-u.ac.jp

  • ↵3 Abbreviations used in this paper: DC, dendritic cell; LP, lamina propria; Treg, regulatory T cells; MLN, mesenteric lymph node; SP, splenic; PP, Peyer’s patch; IEC, intestinal epithelial cell; PLN, peripheral lymph node; pAb, polyclonal Ab.

  • Received December 8, 2008.
  • Accepted August 20, 2009.

References

  1. ↵
    Coombes, J. L., F. Powrie. 2008. Dendritic cells in intestinal immune regulation. Nat. Rev. Immunol. 8: 435-446.
    OpenUrlCrossRefPubMed
  2. ↵
    Tsuji, N. M., A. Kosaka. 2008. Oral tolerance: intestinal homeostasis and antigen-specific regulatory T cells. Trends Immunol. 29: 532-540.
    OpenUrlCrossRefPubMed
  3. ↵
    Worbs, T., U. Bode, S. Yan, M. W. Hoffmann, G. Hintzen, G. Bernhardt, R. Forster, O. Pabst. 2006. Oral tolerance originates in the intestinal immune system and relies on antigen carriage by dendritic cells. J. Exp. Med. 203: 519-527.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Roncarolo, M. G., M. K. Levings, C. Traversari. 2001. Differentiation of T regulatory cells by immature dendritic cells. J. Exp. Med. 193: F5-9.
    OpenUrlFREE Full Text
  5. ↵
    Akbari, O., R. H. DeKruyff, D. T. Umetsu. 2001. Pulmonary dendritic cells producing IL-10 mediate tolerance induced by respiratory exposure to antigen. Nat. Immunol. 2: 725-731.
    OpenUrlCrossRefPubMed
  6. ↵
    Li, M. O., S. Sanjabi, R. A. Flavell. 2006. Transforming growth factor-β controls development, homeostasis, and tolerance of T cells by regulatory T cell-dependent and -independent mechanisms. Immunity 25: 455-471.
    OpenUrlCrossRefPubMed
  7. ↵
    Jang, M. H., N. Sougawa, T. Tanaka, T. Hirata, T. Hiroi, K. Tohya, Z. Guo, E. Umemoto, Y. Ebisuno, B. G. Yang, et al 2006. CCR7 is critically important for migration of dendritic cells in intestinal lamina propria to mesenteric lymph nodes. J. Immunol. 176: 803-810.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Mowat, A. M.. 2003. Anatomical basis of tolerance and immunity to intestinal antigens. Nat. Rev. Immunol. 3: 331-341.
    OpenUrlCrossRefPubMed
  9. ↵
    Sun, C. M., J. A. Hall, R. B. Blank, N. Bouladoux, M. Oukka, J. R. Mora, Y. Belkaid. 2007. Small intestine lamina propria dendritic cells promote de novo generation of Foxp3 Treg cells via retinoic acid. J. Exp. Med. 204: 1775-1785.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Benson, M. J., K. Pino-Lagos, M. Rosemblatt, R. J. Noelle. 2007. All-trans retinoic acid mediates enhanced Treg cell growth, differentiation, and gut homing in the face of high levels of co-stimulation. J. Exp. Med. 204: 1765-1774.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Coombes, J. L., K. R. Siddiqui, C. V. Arancibia-Carcamo, J. Hall, C. M. Sun, Y. Belkaid, F. Powrie. 2007. A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-β and retinoic acid-dependent mechanism. J. Exp. Med. 204: 1757-1764.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Uematsu, S., K. Fujimoto, M. H. Jang, B. G. Yang, Y. J. Jung, M. Nishiyama, S. Sato, T. Tsujimura, M. Yamamoto, Y. Yokota, et al 2008. Regulation of humoral and cellular gut immunity by lamina propria dendritic cells expressing Toll-like receptor 5. Nat. Immunol. 9: 769-776.
    OpenUrlCrossRefPubMed
  13. ↵
    Munn, D. H., M. D. Sharma, J. R. Lee, K. G. Jhaver, T. S. Johnson, D. B. Keskin, B. Marshall, P. Chandler, S. J. Antonia, R. Burgess, et al 2002. Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 297: 1867-1870.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Mellor, A. L., D. H. Munn. 2004. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat. Rev. Immunol. 4: 762-774.
    OpenUrlCrossRefPubMed
  15. ↵
    Belladonna, M. L., U. Grohmann, P. Guidetti, C. Volpi, R. Bianchi, M. C. Fioretti, R. Schwarcz, F. Fallarino, P. Puccetti. 2006. Kynurenine pathway enzymes in dendritic cells initiate tolerogenesis in the absence of functional IDO. J. Immunol. 177: 130-137.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    van der Marel, A. P., J. N. Samsom, M. Greuter, L. A. van Berkel, T. O'Toole, G. Kraal, R. E. Mebius. 2007. Blockade of IDO inhibits nasal tolerance induction. J. Immunol. 179: 894-900.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Mellor, A. L., J. Sivakumar, P. Chandler, K. Smith, H. Molina, D. Mao, D. H. Munn. 2001. Prevention of T cell-driven complement activation and inflammation by tryptophan catabolism during pregnancy. Nat. Immunol. 2: 64-68.
    OpenUrlCrossRefPubMed
  18. ↵
    Munn, D. H., M. D. Sharma, D. Hou, B. Baban, J. R. Lee, S. J. Antonia, J. L. Messina, P. Chandler, P. A. Koni, A. L. Mellor. 2004. Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes. J. Clin. Invest. 114: 280-290.
    OpenUrlCrossRefPubMed
  19. ↵
    Popov, A., Z. Abdullah, C. Wickenhauser, T. Saric, J. Driesen, F. G. Hanisch, E. Domann, E. L. Raven, O. Dehus, C. Hermann, et al 2006. Indoleamine 2,3-dioxygenase-expressing dendritic cells form suppurative granulomas following Listeria monocytogenes infection. J. Clin. Invest. 116: 3160-3170.
    OpenUrlCrossRefPubMed
  20. ↵
    Ueno, A., S. Cho, L. Cheng, J. Wang, S. Hou, H. Nakano, P. Santamaria, Y. Yang. 2007. Transient upregulation of indoleamine 2,3-dioxygenase in dendritic cells by human chorionic gonadotropin downregulates autoimmune diabetes. Diabetes 56: 1686-1693.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Grohmann, U., C. Orabona, F. Fallarino, C. Vacca, F. Calcinaro, A. Falorni, P. Candeloro, M. L. Belladonna, R. Bianchi, M. C. Fioretti, P. Puccetti. 2002. CTLA-4-Ig regulates tryptophan catabolism in vivo. Nat. Immunol. 3: 1097-1101.
    OpenUrlCrossRefPubMed
  22. ↵
    Fallarino, F., C. Vacca, C. Orabona, M. L. Belladonna, R. Bianchi, B. Marshall, D. B. Keskin, A. L. Mellor, M. C. Fioretti, U. Grohmann, P. Puccetti. 2002. Functional expression of indoleamine 2,3-dioxygenase by murine CD8α+ dendritic cells. Int. Immunol. 14: 65-68.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Read, S., V. Malmstrom, F. Powrie. 2000. Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25+CD4+ regulatory cells that control intestinal inflammation. J. Exp. Med. 192: 295-302.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Alegre, M. L., K. A. Frauwirth, C. B. Thompson. 2001. T-cell regulation by CD28 and CTLA-4. Nat. Rev. Immunol. 1: 220-228.
    OpenUrlCrossRefPubMed
  25. ↵
    Fallarino, F., U. Grohmann, K. W. Hwang, C. Orabona, C. Vacca, R. Bianchi, M. L. Belladonna, M. C. Fioretti, M. L. Alegre, P. Puccetti. 2003. Modulation of tryptophan catabolism by regulatory T cells. Nat. Immunol. 4: 1206-1212.
    OpenUrlCrossRefPubMed
  26. ↵
    Iellem, A., M. Mariani, R. Lang, H. Recalde, P. Panina-Bordignon, F. Sinigaglia, D. D'Ambrosio. 2001. Unique chemotactic response profile and specific expression of chemokine receptors CCR4 and CCR8 by CD4+CD25+ regulatory T cells. J. Exp. Med. 194: 847-853.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Wing, K., Y. Onishi, P. Prieto-Martin, T. Yamaguchi, M. Miyara, Z. Fehervari, T. Nomura, S. Sakaguchi. 2008. CTLA-4 control over Foxp3+ regulatory T cell function. Science 322: 271-275.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Yamamoto, M., K. Fujihashi, K. Kawabata, J. R. McGhee, H. Kiyono. 1998. A mucosal intranet: intestinal epithelial cells down-regulate intraepithelial, but not peripheral, T lymphocytes. J. Immunol. 160: 2188-2196.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    Braun, D., R. S. Longman, M. L. Albert. 2005. A two-step induction of indoleamine 2,3-dioxygenase (IDO) activity during dendritic-cell maturation. Blood 106: 2375-2381.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Mellor, A. L., B. Baban, P. Chandler, B. Marshall, K. Jhaver, A. Hansen, P. A. Koni, M. Iwashima, D. H. Munn. 2003. Cutting edge: induced indoleamine 2,3-dioxygenase expression in dendritic cell subsets suppresses T cell clonal expansion. J. Immunol. 171: 1652-1655.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    Munn, D. H., M. D. Sharma, A. L. Mellor. 2004. Ligation of B7-1/B7-2 by human CD4+ T cells triggers indoleamine 2,3-dioxygenase activity in dendritic cells. J. Immunol. 172: 4100-4110.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Guo, Z., M. H. Jang, K. Otani, Z. Bai, E. Umemoto, M. Matsumoto, M. Nishiyama, M. Yamasaki, S. Ueha, K. Matsushima, et al 2008. CD4+CD25+ regulatory T cells in the small intestinal lamina propria show an effector/memory phenotype. Int. Immunol. 20: 307-315.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Verbovetski, I., H. Bychkov, U. Trahtemberg, I. Shapira, M. Hareuveni, O. Ben-Tal, I. Kutikov, O. Gill, D. Mevorach. 2002. Opsonization of apoptotic cells by autologous iC3b facilitates clearance by immature dendritic cells, down-regulates DR and CD86, and up-regulates CC chemokine receptor 7. J. Exp. Med. 196: 1553-1561.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Alexander, A. M., M. Crawford, S. Bertera, W. A. Rudert, O. Takikawa, P. D. Robbins, M. Trucco. 2002. Indoleamine 2,3-dioxygenase expression in transplanted NOD islets prolongs graft survival after adoptive transfer of diabetogenic splenocytes. Diabetes 51: 356-365.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Gurtner, G. J., R. D. Newberry, S. R. Schloemann, K. G. McDonald, W. F. Stenson. 2003. Inhibition of indoleamine 2,3-dioxygenase augments trinitrobenzene sulfonic acid colitis in mice. Gastroenterology 125: 1762-1773.
    OpenUrlCrossRefPubMed
  36. ↵
    Williams, C. A., R. A. Harry, J. D. McLeod. 2008. Apoptotic cells induce dendritic cell-mediated suppression via interferon-γ-induced IDO. Immunology 124: 89-101.
    OpenUrlCrossRefPubMed
  37. ↵
    Forster, R., A. C. Davalos-Misslitz, A. Rot. 2008. CCR7 and its ligands: balancing immunity and tolerance. Nat. Rev. Immunol. 8: 362-371.
    OpenUrlCrossRefPubMed
  38. ↵
    Spahn, T. W., H. L. Weiner, P. D. Rennert, N. Lugering, A. Fontana, W. Domschke, T. Kucharzik. 2002. Mesenteric lymph nodes are critical for the induction of high-dose oral tolerance in the absence of Peyer’s patches. Eur. J. Immunol. 32: 1109-1113.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section