Reverse signaling through GITR ligand enables dexamethasone to activate IDO in allergy


Glucocorticoid-induced tumor necrosis factor receptor (GITR) on T cells and its natural ligand, GITRL, on accessory cells contribute to the control of immune homeostasis. Here we show that reverse signaling through GITRL after engagement by soluble GITR initiates the immunoregulatory pathway of tryptophan catabolism in mouse plasmacytoid dendritic cells, by means of noncanonical NF-κB–dependent induction of indoleamine 2,3-dioxygenase (IDO). The synthetic glucocorticoid dexamethasone administered in vivo activated IDO through the symmetric induction of GITR in CD4+ T cells and GITRL in plasmacytoid dendritic cells. The drug exerted IDO-dependent protection in a model of allergic airway inflammation. Modulation of tryptophan catabolism via the GITR-GITRL coreceptor system might represent an effective therapeutic target in immune regulation. Induction of IDO could be an important mechanism underlying the anti-inflammatory action of corticosteroids.


The immune system exploits a system of bidirectional communication, termed 'reverse' or 'outside-in' signaling, whereby pairs of coreceptors on adjacent cells engage in a cross-talk by reciprocally acting as ligands and receptors. This cooperative form of intercellular communication, in which ligands work heterodoxly as receptors, typically involves members of the tumor necrosis factor (TNF) family1,2. It also applies to the triad of coreceptors—cytotoxic T lymphocyte–associated antigen 4 (CTLA-4), CD28 and B7—enabling bidirectional and univocal conditioning of T lymphocytes and dendritic cells (DCs)3,4,5. As a result, mouse and human DCs respond to CTLA-4 engagement of their B7 molecules with activation of the regulatory pathway of tryptophan catabolism. This pathway, initiated by the enzyme indoleamine 2,3-dioxygenase (IDO), is now considered to be one of the contact-dependent effector mechanisms of natural regulatory T (Treg) cells that express surface CTLA-4 (ref. 6).

In addition to CTLA-4, Treg cells possess the glucocorticoid-inducible TNF receptor (GITR)7,8. First found on a T hybridoma treated with the steroid dexamethasone, this receptor is capable of inhibiting T-cell receptor–induced apoptosis9. Its natural ligand, GITRL, belongs to the TNF superfamily and is present on a variety of cells, including mature and immature DCs (refs. 10,11). The engagement of this receptor by GITRL can reverse the suppression by Treg cells and costimulate effector T cells12,13,14,15,16. This is consistent with the general view that the cross-talk between TNF family members typically results in the activation or enhancement of an immune response. Recent studies with an agonistic antibody have revealed, however, that activation of 4-1BB—a member of the TNF receptor family—blocks the progression of experimental autoimmunity. The effect is dependent on tryptophan catabolism and requires the participation of specific subsets of Treg cells and DCs (ref. 17).

Plasmacytoid DCs (pDCs) represent a specialized cell population that produces large amounts of type I interferons (IFNs) in response to viruses18. The ability of pDCs to secrete type I IFNs depends on cellular sensors that promptly detect the presence of DNA and RNA viruses19. pDCs participate in the defense against viruses and in the regulation of immunity, including the induction and maintenance of tolerance20. Stimulation of pDCs with the tolerogenic ligands CTLA-4–immunoglobulin (CTLA-4–Ig)21 and CD200-Ig (ref. 22), or with CpG-rich oligodeoxynucleotides (ODNs)23, initiates the regulatory pathway of tryptophan catabolism mediated by IDO.

By using a soluble form of GITR (GITR-Ig), we demonstrate here that pDCs possess GITRL and that reverse signaling through GITRL results in noncanonical NF-κB activation and the onset of IDO-dependent immune regulation. Dexamethasone administered in vivo activated IDO through the concordant induction of GITR in CD4+ T cells and GITRL in pDCs. IDO activation by the glucocorticoid contributed to protection against allergic bronchopulmonary aspergillosis (ABPA). The GITRL-dependent modulation of tryptophan catabolism could represent an important mechanism of action of glucocorticoids, both as physiologic mediators and as therapeutic agents.


GITR-Ig activates IDO in pDCs

In a preliminary series of experiments combining FACS analysis with gene silencing by small interfering RNA (siRNA) technology, we obtained evidence that a fusion protein of mouse GITR and IgG3 Fc (GITR-Ig) selectively binds GITRL-expressing pDCs (Fig. 1a–c and Supplementary Note online). We asked whether GITRL engagement by GITR-Ig would affect the function of pDCs in a skin test assay measuring the tolerogenic potential of those cells22,23,24. We injected otherwise immunogenic, peptide-pulsed conventional CD8 DCs alone or in combination with pDCs (a fraction constituting 5% of the final cell mixture), either untreated or treated with CpG ODN, CTLA-4–Ig, GITR-Ig or control Ig-Cγ3 (a fusion protein that selectively contains the IgG3 Fc domain5,24) (Fig. 2a). The priming ability of the CD8 DCs was unaffected by the presence of either untreated or Ig-Cγ3–treated pDCs (ref. 24). CpG, CTLA-4–Ig and GITR-Ig were all capable of suppressive effects in pDCs that would preclude host priming by cotransferred CD8 DCs.

Figure 1: Splenic pDCs constitutively express GITRL, which specifically recognizes soluble GITR.

(a) Kinetic RT-PCR analysis of GITRL-encoding Tnfsf18 expression in pDCs treated with Tnfsf18-specific siRNA. Control cells were treated with negative control (nc) siRNA. Transcriptional expression of the 4-1BBL–encoding Tnfsf9 gene was assayed as a specificity control. (b,c) Comparative FACS analyses of GITRL-specific antibody and GITR-Ig binding to pDCs. CD11c+mPDCA-1+ cells were transfected with nc or Tnfsf18-specific siRNA to be assayed, on day 3 of transfection, for binding of antibody to GITRL or GITR-Ig and for expression of the 120G8 marker. In b, cells were stained with 120G8-Alexa 488 and PE-labeled antibody to mouse GITRL or an isotype control. In c, cells were treated with 1 μg/ml GITR-Ig or control Ig-Cγ3 for 30 min on ice, and then reacted with PE-labeled antibody to mouse IgG3; these cells were also stained with 120G8-Alexa 488. In b and c, the percentage of cells present in the upper right quadrant is indicated. Controls (not shown) included an isotype control for 120G8 staining in b, as well as PE-labeled antibodies to mouse IgG3 used alone and isotype control for 120G8 in c. FACS analyses are representative of at least three different experiments.

Figure 2: GITRL engagement by GITR-Ig confers suppressive properties on pDCs that require IDO expression and function.

(a) CD11c+ DCs were fractionated according to CD8 or mPDCA-1 expression, pulsed with NRP-A7, and transferred into recipient mice to be assayed for skin reactivity to the eliciting peptide. The CD8 DC fraction was used alone or in combination with 5% pDCs conditioned by CpG, CTLA-4–Ig, GITR-Ig or control Ig-Cγ3. (b) The pDC fraction was also used after transfection with negative control (nc), Tnfsf18-specific siRNA or Indo-specific siRNA. In a and b, the development of NRP-A7–specific skin reactivity was assessed at 2 weeks after cell transfer. Data are mean ± s.d. of replicate experiments (in a, n = 4 and *P < 0.005; in b, n = 3 and *P < 0.01). (c) IDO protein expression was assessed by immunoblot analysis, using an IDO-specific antibody. pDCs were treated in vitro with GITR-Ig, CTLA-4–Ig or Ig-Cγ3. Results are from one experiment representative of four. The mean fold changes (± s.d.) in the four experiments (that is, ratio of Ig-Cγ3, GITR-Ig or CTLA-4–Ig treatment to no treatment, using normalized densitometry values) were 1.1 ± 0.2, 4.0 ± 0.8 and 4.4 ± 0.9 for the respective treatments in that order. (d) Functional IDO activity in response to GITR-Ig was measured in vitro in terms of the ability to metabolize tryptophan to kynurenine, using pDCs transfected with nc, Tnfsf18-specific siRNA or Indo-specific siRNA. Kynurenine concentrations were measured in supernatants. Results are the mean ± s.d. of three experiments. *P < 0.05.

We subjected the pDCs to silencing of Tnfsf18 (also known as Gitrl) or Indo before treatment with GITR-Ig and mixing with the CD8 DCs. We also assayed transfected pDCs treated with CTLA-4–Ig in place of GITR-Ig (Fig. 2b). Indo silencing, but not silencing of the GITRL-encoding gene, blocked the suppressive effect of CTLA-4–Ig (ref. 3). Silencing either gene negated the effect of GITR-Ig on pDCs. We also observed the induction of IDO protein and function by GITR-Ig in vitro (Fig. 2c,d and Supplementary Note). Therefore, GITRL engagement by GITR-Ig in pDCs activates a suppressive program that is contingent on IDO.

IFN-α is necessary for GITR-Ig effects

On the basis of preliminary evidence that GITR-Ig causes release of IFNs from pDCs (Fig. 3a and Supplementary Note), we investigated the role of type I and type II IFNs in the IDO-dependent effects of GITR-Ig in vivo. In a skin test assay, we injected peptide-pulsed CD8 DCs alone or in combination with pDCs treated with GITR-Ig, with or without antibodies to IFN-α or antibodies to IFN-γ. As an alternative, we used pDCs from mice deficient in STAT1 (Stat1−/−), IFN-γ (Ifng−/−) or the IFN-α/β receptor (Ifnar−/−). As a control, we assayed Ifnar−/− pDCs after treatment with CTLA-4–Ig (Fig. 3b). pDCs treated with Ig-Cγ3 did not inhibit priming, unlike their counterparts treated with GITR-Ig. IFN-γ neutralization had no effect on the inhibitory activity of GITR-Ig, as opposed to treatment with antibodies to IFN-α or to the genetic deficiency of STAT1 or the IFN-α/β receptor. Ifnar−/− pDCs were fully responsive to CTLA-4–Ig. Thus type I IFNs, and in particular IFN-α, seemed to be required for GITR-Ig to confer a suppressive activity on pDCs. Reverse transcription PCR (RT-PCR) analysis of the genes encoding IFN-α and IFN-β revealed that GITR-Ig mostly induced transcriptional expression of the former gene (Fig. 3c).

Figure 3: IFN-α dependence of IDO induction in pDCs by GITR-Ig.

(a) CD11c+mPDCA-1+ cells were treated with GITR-Ig, CTLA-4–Ig, Ig-Cγ3 or CpG ODN for 24 h, and IFN-α and IFN-γ contents in supernatants were assayed by ELISA. Data are mean ± s.d. (n = 3). *P < 0.01 and **P < 0.001, for treatment versus medium alone. (b) CD8 DCs were pulsed with the HY peptide and transferred into female mice to be assayed for skin reactivity. Cells were used in combination with 5% pDCs from wild-type, Stat1−/−, Ifng−/− or Ifnar−/− mice. The pDCs were treated with Ig-Cγ3, GITR-Ig or CTLA-4–Ig, used singly or in combination with antibodies to IFN-α or antibodies to IFN-γ. Rat IgG (50 μg/ml) was the control treatment for IFN neutralization. n = 3. *P < 0.05. (c) RT-PCR transcript analysis of the genes encoding IFN-α and IFN-β in pDCs exposed to GITR-Ig or control Ig-Cγ3 for 4 h or 24 h. Results are from one experiment representative of three. (d) Kynurenine production (n = 3) was measured in pDCs treated with GITR-Ig in the presence of antibodies to IFN-α or antibodies to IFN-γ. Alternatively, cells were exposed to different concentrations per ml of recombinant IFNs (rIFNs), singly or in combination, in the absence of GITR-Ig. *P < 0.05 for treatment with rIFNs versus control (C); **P < 0.01 for treatment with anti–IFN-α or rIFNs versus rat IgG or C treatment, respectively. P = 0.331 for treatment with antibody to IFN-γ versus treatment with IgG.

We next conducted in vitro experiments using GITR-Ig in combination with antibodies to either IFN-α or IFN-γ, and using recombinant cytokines in the absence of GITR-Ig. Our results provided evidence for a necessary, but not sufficient, role for IFN-α in IDO induction by GITR-Ig (Fig. 3d) and Supplementary Note).

GITR-Ig activates noncanonical NF-κB signaling

NF-κB has an important regulatory role in the IFN-inducible expression of Indo (ref. 25). Although much attention has been focused on the pro-inflammatory signaling of NF-κB, recent data indicate that the two IκB kinase (IKK) complex catalytic subunits IKKα and IKKβ could have opposing roles. IKKα is thought to limit NF-κB activity and control inflammation26, whereas IKKβ is indispensable in the so-called canonical pathway of NF-κB activation. IKKα, therefore, could be pivotal in the 'noncanonical' or 'alternative' NF-κB activation that leads to resolution of the early inflammatory process26,27 and establishment of self-tolerance28. The noncanonical pathway is typically induced by members of the TNF receptor family, and activation of IKKα by the NF-κB–inducing kinase (NIK) results in the processing of p100 to p52 and the consequent formation of p52-RelB dimers, which translocate into the nucleus27.

We examined GITR-Ig for possible activation of noncanonical NF-κB in pDCs, and analyzed the effects of silencing Map3k14 (also known as Nik), Chuk (which encodes IKKα) or Ikbkb (which encodes IKKβ) on the functional expression of IDO by pDCs in vitro and in vivo. GITR-Ig activated the phosphorylation of NIK and IKKα, an effect evident after 5 min for NIK and at 20 min for IKKα (Fig. 4a). We used immunoblot analysis to determine the relative amounts of p100 and p52 in pDCs treated with GITR-Ig for different lengths of time (Supplementary Fig. 1 online). GITR-Ig stimulation resulted in a progressive increase in the relative amounts of p52, with peak levels at 30 min. We next quantified the activation of the NF-κB family, using an ELISA kit specific for mouse p65, p52 and RelB (Fig. 4b). Significant nuclear translocation occurred for p52 and RelB, but not for p65, following 30–60 min of exposure to GITR-Ig. Using specific gene silencing in combination with different approaches (Supplementary Figs. 2 and 3 online and Fig. 4c–e; details in Supplementary Note), we found that lack of functional NIK or IKKα, but not IKKβ, negated the IFN-α–inducing and IDO-enhancing effects of GITR-Ig in vitro as well as the ability of the pDCs to block priming by cotransferred CD8 DCs in vivo. Therefore, signaling through GITRL leads to noncanonical NF-κB induction, which is necessary for IDO function.

Figure 4: GITR-Ig activates noncanonical NF-κB signaling leading to IDO functional expression.

(a) pDCs were treated with GITR-Ig or control Ig-Cγ3 for different lengths of time, and immunoblots of cell lysates were probed with antibodies to pNIK or pIKKα/β, followed by antibodies to NIK, IKKα or IKKβ. The expected migration of pIKKβ is indicated, and the blot is representative of three experiments. (As the anti-pIKK antiserum detects both IKKα and IKKβ, we confirmed detection of pIKKβ using cells treated with TNF-α.) (b) Results from an ELISA procedure used to monitor activation of p65, p52 and RelB in nuclear extracts from cells treated with GITR-Ig or control Ig-Cγ3 for different lengths of time. Time 0 indicates untreated cells. Relative activities (A450) are mean ± s.d. of three experiments, each in triplicate. *P < 0.05 and **P < 0.01, for treated versus untreated cells. (c) Kinetic RT-PCR analysis of Nik, IKKα-encoding Chuk and IKKβ-encoding Ikbkb expression in pDCs treated with specific siRNAs. Data are from one experiment representative of three. Control cells were treated with negative control (nc) siRNA. (d) Functional IDO in response to GITR-Ig was measured in vitro in terms of tryptophan conversion to kynurenine, using pDCs transfected with nc, Nik, Chuk or Ikbkb siRNAs. Kynurenine levels were measured in supernatants (n = 3). *P < 0.05 and **P = 0.01. (e) CD11c+ DCs were fractionated according to CD8 or mPDCA-1 expression, pulsed with NRP-A7, and transferred into recipient mice to be assayed for skin reactivity to the eliciting peptide. The CD8 DC fraction was used in combination with 5% pDCs treated with GITR-Ig or Ig-Cγ3. The pDCs had been transfected with nc, Nik, Chuk or Ikbkb siRNAs. n = 4. *P < 0.005, for experimental versus control footpads.

Dexamethasone causes GITR-dependent induction of IDO

We asked whether the treatment of mice with dexamethasone would induce changes in the levels of GITRL and/or GITR (ref. 29) that might affect the pattern of IDO expression by pDCs. We used wild-type and Tnfrsf18−/− (Gitr−/−) mice treated with the drug as donors of splenic CD4+ T cells and mPDCA-1+120G8+ pDCs. We assayed the CD4+ T cell fraction for expression of GITR and examined the pDCs for expression of GITRL (Supplementary Fig. 4 online). Dexamethasone doubled the fraction of GITR+ CD4+ cells in wild-type mice, and the percentage of pDCs expressing GITRL increased to a similar extent in both wild-type and GITR-deficient mice. The expression of CD25 and CTLA-4 in CD4+ T cells was unaffected. Concentration-response assays of IFN-α and IFN-γ production by pDCs showed that the IFN-α response to GITR-Ig in vitro was increased by treating the prospective pDC donors with the drug (Supplementary Fig. 5 online). This indicated that the twofold increase in GITR and GITRL expressions in vivo by dexamethasone may have a physiologically relevant functional effect. Additional studies demonstrated that the drug enhanced tryptophan catabolism in pDCs by means of a mechanism relying on GITR, but not CTLA-4, expression by T cells and independent of Toll-like receptor (TLR)-9 signaling and the intracellular adaptor protein myeloid differentiation factor 88 (MyD88) pathway (Fig. 5a,b and Supplementary Note).

Figure 5: Dexamethasone causes GITR-dependent expression of IDO protein and function.

Wild-type (WT) and GITR-deficient Tnfrsf18−/− mice received daily treatments with 2.5 mg/kg dexamethasone (dex) or vehicle alone (days 0–4). On day 5, mice served as donors of mPDCA-1+120G8+ pDCs, which were purified by fluorescence-activated sorting from splenic CD11c+ cells. (a) IDO protein expression was assessed in pDCs from WT, Tnfrsf18−/−, Tlr9−/− and Myd88−/− mice. A group of dex-treated WT donors of pDCs received antibody to CTLA-4 (100 μg/mouse on day −1, followed by 50 μg on days 0 and 1). Hamster IgG was used for control treatment (not shown). The blot is representative of three experiments. (b) Functional IDO was assayed in vitro using the same pDCs as in a. Kynurenine concentrations were measured in supernatants (n = 3). *P < 0.05, dex versus vehicle treatment. (c) CD11c+CD8 DCs from WT mice were pulsed with HY and transferred into recipient mice to be assayed for skin reactivity to the eliciting peptide. The CD8 DC fraction was used in combination with 5% pDCs from dex-treated WT or Tnfrsf18−/− mice. Groups of recipient mice were treated with placebo or 1-MT pellets. A portion of the pDCs from Tnfrsf18−/− mice on dex was treated in vitro with GITR-Ig. Development of HY-specific skin test reactivity was assessed at 2 weeks after cell transfer (n = 5); *P < 0.005.

In vivo, in a skin test assay using the cotransfer model, we injected peptide-pulsed CD8 DCs into recipient hosts either alone or in combination with pDCs from dexamethasone-treated wild-type or GITR-deficient mice (Fig. 5c). Because implants of the IDO inhibitor 1-methyl-tryptophan (1-MT) inhibit tryptophan catabolism in vivo3, we treated groups of recipient mice with placebo or 1-MT pellets. The pDCs from wild-type, but not GITR-deficient, mice treated with the drug manifested an ability to suppress host priming by cotransferred CD8 DCs, and the suppressive effect required functional IDO in the recipients. Although the pDCs from GITR-deficient mice on dexamethasone were not suppressive, they acquired the IDO-dependent suppressive ability when exposed to GITR-Ig in vitro before cotransfer. This demonstrates that dexamethasone in vivo confers suppressive properties on pDCs that are dependent on GITR expression by the host and require the IDO mechanism as a downstream effector system.

IDO induction by dexamethasone protects against ABPA

Allergic bronchopulmonary aspergillosis is a T-helper type 2 (TH2)-sustained allergic condition of the lungs that is responsive to steroid treatment30,31. Experimental models of the disease have been used to demonstrate a pivotal role for Treg cells32, pDCs (ref. 33) and tryptophan catabolism34 in protecting mice from allergic airway inflammation. We evaluated the IDO-dependent effects of dexamethasone on the hypersensitivity response to Aspergillus antigens in the mouse lung. We established bronchial colonization of A. fumigatus after the elicitation of a strong TH2 reactivity, and then treated the mice with dexamethasone and 1-MT or placebo in order to assess parameters of allergic airway inflammation. At 1 week of colonization, the number of eosinophils and the mucin secretion in the bronchoalveolar lavage (BAL) fluid, and the amounts of circulating IgE antibody and lung hydroxyproline—all indicative of A. fumigatus respiratory allergy32—were decreased in dexamethasone-treated mice relative to mice that were untreated or cotreated with 1-MT (Fig. 6a). In the last group, histopathology confirmed the presence of extensive peribronchial infiltrates of mononuclear cells and eosinophils, goblet cell hyperplasia and mucus deposition in the airways (Fig. 6b and Supplementary Note).

Figure 6: IDO-dependent effects of dexamethasone in allergic airway inflammation.

Bronchial colonization with A. fumigatus was established in TH2-primed mice that were treated with dexamethasone (dex) and either 1-MT or placebo pellets, to be assayed for A. fumigatus allergy at 1 week. Sham-sensitized, colonized mice were present in all assays (control), and so were sensitized and colonized mice on dex vehicle alone (vehicle) or on 1-MT alone (1-MT). (a) The abundance of eosinophils (eos), macrophages (mac), lymphocytes (lym) and neutrophils (neu) was assessed in BAL fluid, along with mucin concentrations. IgE antibody and hydroxyproline were determined in sera and lung homogenates, respectively. Secreted cytokines were measured in BAL fluid. Thoracic lymph node (TLN) CD4+ T cells were assayed for cytokine production in vitro in response to antibodies to CD3. Data are mean ± s.d. of three experiments. *P < 0.05 and **P < 0.005. (b) Paraffin-embedded lung sections were stained with hematoxylin and eosin (H&E) to evaluate inflammatory cells, or with periodic acid-Schiff (PAS) to visualize globet cells. Data are representative of three independent experiments. (c) Gata3 and Foxp3 transcripts were evaluated in CD4+ T cells from thoracic lymph nodes. Sorted CD4+ lymphocytes (5 × 105/ml) were activated with soluble antibodies to CD3 for 24 h. Gata3 and Foxp3 mRNAs were quantified by real-time PCR using Gapdh normalization. Data (mean ± s.d. of four experiments) are presented as normalized transcript expression in the samples relative to normalized transcript expression in the control culture (cells from sham-sensitized mice; that is, fold change = 1, dotted line).

We analyzed the patterns of TH1, TH2 and Treg-associated cytokine production in BAL fluid and thoracic lymph nodes. We found a profile typical of A. fumigatus allergy in mice that were untreated or treated with the drug and 1-MT, characterized by a dominant production of interleukin (IL)-5, IL-4 and IL-13, and a reduced expression of TH1-associated IFN-γ (refs. 3234 and Fig. 6a). Again, the TH2-dependent allergic phenotype was greatly attenuated by dexamethasone, which enhanced the production of IL-10, a marker of protective Treg activity in A. fumigatus allergy34.

The cytokine pattern was consistent with PCR assessment of Gata3 and Foxp3 transcripts in thoracic lymph node CD4+ T cells (Fig. 6c). In mice treated with dexamethasone but not 1-MT, the TH2 transcription factor GATA3 was downregulated and the Treg lineage specification factor Foxp3 was enhanced. These data demonstrate that the steroid decreases the exacerbating TH2 responses in ABPA by inhibiting the expansion and activation of TH2 cells and upregulates Foxp3 expression by means of mechanisms that require tryptophan catabolism.


GITR functions as a regulator of physiologic and pathologic immune responses by costimulating T-effector cells and abrogating the suppressive effects of Treg cells12,13,14,15,16. Although these cells do not have a unique cell surface phenotype, a constellation of cell surface proteins, including CD25, CTLA-4 and GITR itself, has enabled investigators to purify Treg cells and to demonstrate their function in vivo and in vitro. However, the functional importance of these markers to overall Treg activity in vivo remains unclear35. Recent data support a model in which Treg cells inhibit the activation, differentiation and survival of pathogenic T cells through enhanced tryptophan catabolism consequent to reverse signaling in DCs, which respond to inhibitory ligands and cytokines expressed by Treg cells3,4,22,24. Paradigmatic in this regard may be the case of CTLA-4: this tolerogenic ligand signals DCs through B7, which also serves as a receptor for immunostimulatory signaling by CD28 (ref. 5).

GITRL is expressed in endothelial and antigen-presenting cells, including DCs (refs. 7,10,11). We focused our attention on pDCs for several reasons, including our own preliminary observation of a higher expression of the Tnfsf18 transcript in pDCs than in different types of conventional DCs. Also, because pDCs produce large amounts of cytokines, particularly type I IFNs, they regulate inflammation and link innate responses with adaptive immunity18,19 and are functionally related to Treg cells20. Finally, although pDCs have been credited with a unique ability to produce IDO in experimental settings6, they are not tolerogenic per se, and multiple ligands and cytokines contribute to the expression of a suppressive phenotype by these cells21,22,24,33. We found that reverse signaling through GITRL acts on pDCs to activate IDO in a fashion contingent on noncanonical NF-κB pathway induction. This suggests a potential biological relevance of reverse signaling through GITRL in pDCs as a means of balancing the immunoenhancing and costimulatory effects of GITR activation on T cells.

Traditionally recognized for its role in infection, pregnancy, transplantation, autoimmunity and neoplasia6,36,37, the IDO mechanism has revealed an unexpected potential role in the control of inflammation17,38, allergy and allergic airway inflammation32,39, all conditions in which pDCs could have a protective function40. Some of the IDO-dependent effects are initiated by CpG ODN treatment and rely on cell signaling through TLR9 (refs. 22,39,41,42), whose activation typically results in type I IFN release by the pDCs (ref. 19). Type I IFNs activate noncanonical NF-κB signaling43. Although it is possible that IFN-α contributed to noncanonical NF-κB activation in our setting, phosphorylation of NIK and increased formation of p52 were already evident within minutes of GITR-Ig exposure and were likewise observable with pDCs from Ifnar−/− mice (Supplementary Fig. 6 online). Thus, although IFN-α and STAT1 could be instrumental in enabling the functional responsiveness of pDCs to GITR-Ig, and IFN-α could contribute to a sustained NF-κB activation, the induction of the noncanonical pathway by GITR-Ig was upstream of IFN-α production.

Reciprocally, transcriptional regulation of type I IFN-encoding genes in pDCs is controlled mainly by the IFN regulatory factor (IRF)-3 and IRF7 (ref. 19), and the Irf3 promoter contains a noncanonical NF-κB binding site recognized by p52-RelB dimers44. It is therefore possible that noncanonical NF-κB contributed to IFN-α production in response to GITR-Ig. In a manner similar to IFN-γ, but even more potently, IFN-α induces Indo expression and function in pDCs. Thus autocrine IFN-α could in principle mediate at least a portion of the upregulation of Indo in cells treated with GITR-Ig. However, IFN-α cannot account per se for the full extent of GITR-Ig effects, as demonstrated by our experiments of tryptophan conversion in vitro by the recombinant cytokine used alone or in combination with IFN-γ. In these experiments we used cytokine concentrations similar to those present in supernatants of pDCs treated with GITR-Ig. The Indo promoter contains a putative noncanonical NF-κB half-site (GGGAGA) recognized by p52-RelB dimers at position −3,566 (ref. 44). In addition, IDO activity is critically regulated at the post-translational level by a set of genes that are comodulated by inhibitory ligands26.

Dexamethasone treatment in vivo increased the amount of GITR in CD4+ T cells and of GITRL in pDCs. Treatment also conferred immunoregulatory properties on pDCs that were dependent on GITR expression by the host and required functional IDO in vivo. Thus it seems likely that through the synergistic effects on T-cell expression of GITR and pDC expression of GITRL, dexamethasone acts on the GITR-GITRL coreceptor system to modulate IDO activation through noncanonical NF-κB signaling. Such a mechanism would be consistent with a unitary role of pDCs (refs. 33,40), noncanonical NF-κB (refs. 26,28) and IDO function (refs. 17,32,38,39) in controlling inflammatory pathology and allergy.

Previous studies have identified a mechanism that could, potentially, prevent excessive inflammation during infection. Such a mechanism intrinsically links maturing pDCs to the generation of IL-10–producing Treg cells through TLR-dependent and TLR-independent pathways45,46. It is unknown whether the TLR-independent mechanism, CD40 signaling, activates noncanonical NF-κB (ref. 27) and IDO (ref. 47) in the setting of infection-driven inflammation. Because IDO is involved in the peripheral generation of Treg cells48, it is possible that multiple ligands45,46, glucocorticoids49 and pDCs (ref. 20) favor Treg generation through pathways converging on noncanonical NF-κB and tryptophan catabolism. Notably, CD40 signaling, which activates both canonical and noncanonical NF-κB (ref. 27), has been found, as one would expect, to diametrically modulate IDO depending on environmental factors47,50.

TRL9 modulation of IDO is effective not only in the prevention and treatment of animal models of allergic disorders23,39,41,42, but also in maintaining a physiologic state of protective tolerance to ubiquitous fungal aeroantigens32,33,34. Allergic bronchopulmonary aspergillosis, a TH2-dependent hypersensitivity lung disease due to bronchial colonization of A. fumigatus, affects 1–2% of asthmatic patients and 7–9% of cystic fibrosis patients. Although the organism can cause allergic disease in otherwise healthy individuals and devastating disease in immunosuppressed individuals, ABPA is a hypersensitivity response to A. fumigatus antigens in the lung and is distinct from other forms of A. fumigatus pulmonary disease30. Oral steroids emerged as a treatment for ABPA after a series of uncontrolled studies showed therapeutic effects in patients on prolonged oral glucocorticoids, and oral steroids remain the mainstay of treatment31. In an experimental model of ABPA in which the combined effects of Treg cells32, pDCs (ref. 33) and tryptophan catabolism34 mediate protection, we found that dexamethasone inhibited TH2 responses and allergy, and induced Foxp3 expression in CD4+ T cells by means of mechanisms dependent on tryptophan catabolism. This further supports the idea of a link between glucocorticoids, pDC expression of GITRL, and Treg activity within a positive feedback loop whereby GITR+ Treg cells would expand their own population through IDO (refs. 20,33,39,48,49).

In conclusion, the present study suggests that reverse signaling through GITRL in pDCs could represent a physiological means of activating noncanonical NF-κB signaling and IDO. This would balance the immunostimulatory effects of GITR activation on T cells. At the same time, modulation of tryptophan catabolism through the GITR-GITRL coreceptor system might represent an important mechanism of action of anti-inflammatory corticosteroids and a new therapeutic approach in several pathologic conditions. The ability of glucocorticoids to exert IDO-dependent protection in a model of allergic airway inflammation could provide new clues to an improved understanding of the complex action of these drugs in human respiratory allergies and asthma.


Mice, DC purification and reagents.

Eight- to ten-week-old female BALB/c (H-2d) and C57BL/6 (H-2b) mice were obtained from Charles River Breeding Laboratories. Female mice deficient for the IFN-α/β receptor (Ifnar−/−)22,24 or GITR (Tnfrsf18−/−)15, both on a 129 Sv/Ev (H-2b) background, were generated as described. Mice homozygous for the STAT1 (Stat1−/−)3, IFN-γ (Ifng−/−)3, TLR9 (Tlr9−/−)33 and MyD88 (Myd88−/−)5 targeted mutation, raised on the C57BL/6 background, have also been previously described. All in vivo studies were in compliance with national (Italian Parliament DL 116/92) and Perugia University Animal Care and Use Committee guidelines.

We purified splenic DCs by magnetic-activated sorting using CD11c MicroBeads and MidiMacs (Miltenyi Biotec)3,22. For positive selection of mPDCA-1+ pDCs, we fractionated CD11c+ cells using mPDCA-1 MicroBeads (Miltenyi Biotec). More than 95% of the mPDCA-1+ cells were stained by the 120G8 marker24. Unless otherwise stated, the pDCs used in this study refer to purified CD11c+mPDCA-1+ cells. In selected experiments, we used flow cytometry to purify 120G8+ cells. CD11c+ cells were stained with phycoerythrin (PE)-labeled rat antibody to mPDCA-1 (Miltenyi Biotec) and 120G8-Alexa 488, for 30 min at 4 °C. Then we sorted 120G8+ cells on an EPICS ALTRA (Beckman Coulter)22. For negative selection of conventional CD8 DCs, CD11c+ cells were fractionated by means of CD8α MicroBeads (Miltenyi Biotec)3,22.

We conducted FACS analyses involving rat 120G8-Alexa 488 and PE-labeled goat antibodies to mouse IgG3 (Southern Biotechnology) as described22,24. We analyzed surface expression of CD25 and total expression of CTLA-4 in permeabilized cells as described48. We obtained PE-labeled monoclonal antibodies to mouse GITR (rat IgG2a) from R&D Systems, and antibodies to GITRL (rat IgG1) from eBioscience. Antibody reagents for ELISA determinations of IFN-α and IFN-γ have previously been described5,24. We used neutralizing XMG1.2 (rat IgG2a) to IFN-γ and neutralizing RMMA-1 (rat IgG1) to IFN-α at 10 μg/ml and 50 μg/ml, respectively4. Antibody to mouse CTLA-4 monoclonal 4F10 (hamster IgG) was as described4. We used an ELISA-based TransAM Flexi NFκB Family Kit (Active Motif) to monitor activity of NF-κB family members. For the immunoblot-based transcription factor assays, we purchased antibodies to phospho-NIK and antibodies to NIK from Santa Cruz Biotechnology. We obtained antibodies to phospho-IKKα/β, antibodies to IKKα, antibodies to IKKβ and antibodies to NF-κB2 p100/p52 from Cell Signaling. To detect primary antibodies, we used antibodies to either goat or rabbit immunoglobulin conjugated to horseradish peroxidase (Pierce).

Construction and expression of GITR-Ig and DC treatments.

These procedures are described in the Supplementary Methods online.

siRNA synthesis and transfection, and PCR.

We performed these procedures as outlined in the Supplementary Methods and as described in ref. 24.

Immunization and skin test assay.

We used a skin test assay to measure class I–restricted delayed-type hypersensitivity in response to intra-footpad challenge with synthetic peptides, as previously described4,5,37. In this assay, purified fractions of conventional CD8 DCs mediate the immunogenic presentation of otherwise poorly immunogenic peptides. However, the addition of a minority fraction of pDCs (as few as 5% of the final cell mixture, see below) treated with tolerogenic ligands or with CpG ODNs (refs. 2224) to a population of CD8 DCs inhibits priming by the latter cells through IDO-dependent mechanisms. For immunization in vivo, we loaded cells with the class I H-2Kd–restricted NRP-A7 (KYNKANAFL) peptide in vitro (5 μM, 2 h at 37 °C), before irradiation and intravenous injection into BALB/c recipient hosts5,24. As an alternative, we used the class I H-2Db–restricted HY (WMHHNMDLI) peptide in female C57BL/6 recipients24. We intravenously injected 3 × 105 peptide-loaded CD8 DCs either alone or in combination with 1.5 × 104 peptide-loaded pDCs. At 2 weeks, we measured the response to intra-footpad challenge with the eliciting peptide. The results are expressed as the increase in footpad weight of peptide-injected footpads over that of the respective, vehicle-injected counterparts.

IDO expression and functional analysis.

We examined IDO induction and function as described4. For details, see Supplementary Methods.

Allergic bronchopulmonary aspergillosis, collection and analysis of BAL fluid, and respiratory allergy phenotype.

The induction and evaluation of the TH2-driven hypersensitivity response to A. fumigatus antigens in the mouse lung have previously been described in detail32,34. Briefly, we sensitized BALB/c mice by the concomitant intraperitoneal (100 μg) and subcutaneous (100 μg) administration of A. fumigatus culture filtrate extract, followed 1 week later by the intranasal instillation of 20 μg of the extract. After an additional 1 week, we induced bronchial colonization of A. fumigatus by administering resting conidia intratracheally (1 × 107). After 1 week, we evaluated the mice for parameters of allergic airway inflammation. Treatments included dexamethasone administered intraperitoneally at 2.5 mg/kg (for five consecutive days, commencing on the day of bronchial colonization) to mice with 1-MT or placebo implants. The analysis of respiratory allergy phenotype is described in the Supplementary Methods.

Statistical analysis.

In the in vivo skin test assay, we performed the statistical analysis using two-tailed paired Student's t-test, by comparing the mean weight of experimental footpads with that of control, saline-injected counterparts4,5,37. Data are expressed as the mean ± s.d. of 3–5 experiments, with at least 6 mice per group per experiment, which yielded a power of at least 80% with an α-level of 0.05 as computed by power analysis. We used a Student's t-test to analyze the results of the in vitro studies; data from these are expressed as the mean ± s.d. of at least three independent experiments.

Note: Supplementary information is available on the Nature Medicine website.


  1. 1

    van Essen, D., Kikutani, H. & Gray, D. CD40 ligand-transduced co-stimulation of T cells in the development of helper function. Nature 378, 620–623 (1995).

    CAS  Article  Google Scholar 

  2. 2

    Eissner, G., Kolch, W. & Scheurich, P. Ligands working as receptors: reverse signaling by members of the TNF superfamily enhance the plasticity of the immune system. Cytokine Growth Factor Rev. 15, 353–366 (2004).

    CAS  Article  Google Scholar 

  3. 3

    Grohmann, U. et al. CTLA-4–Ig regulates tryptophan catabolism in vivo. Nat. Immunol. 3, 1097–1101 (2002).

    CAS  Article  Google Scholar 

  4. 4

    Fallarino, F. et al. Modulation of tryptophan catabolism by regulatory T cells. Nat. Immunol. 4, 1206–1212 (2003).

    CAS  Article  Google Scholar 

  5. 5

    Orabona, C. et al. CD28 induces immunostimulatory signals in dendritic cells via CD80 and CD86. Nat. Immunol. 5, 1134–1142 (2004).

    CAS  Article  Google Scholar 

  6. 6

    Mellor, A.L. & Munn, D.H. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat. Rev. Immunol. 4, 762–774 (2004).

    CAS  Article  Google Scholar 

  7. 7

    Nocentini, G. & Riccardi, C. GITR: a multifaceted regulator of immunity belonging to the tumor necrosis factor receptor superfamily. Eur. J. Immunol. 35, 1016–1022 (2005).

    CAS  Article  Google Scholar 

  8. 8

    Shevach, E.M. & Stephens, G.L. The GITR–GITRL interaction: co-stimulation or contrasuppression of regulatory activity? Nat. Rev. Immunol. 6, 613–618 (2006).

    CAS  Article  Google Scholar 

  9. 9

    Nocentini, G. et al. A new member of the tumor necrosis factor/nerve growth factor receptor family inhibits T cell receptor-induced apoptosis. Proc. Natl. Acad. Sci. USA 94, 6216–6221 (1997).

    CAS  Article  Google Scholar 

  10. 10

    Tone, M. et al. Mouse glucocorticoid-induced tumor necrosis factor receptor ligand is costimulatory for T cells. Proc. Natl. Acad. Sci. USA 100, 15059–15064 (2003).

    CAS  Article  Google Scholar 

  11. 11

    Kim, J.D. et al. Cloning and characterization of GITR ligand. Genes Immun. 4, 564–569 (2003).

    CAS  Article  Google Scholar 

  12. 12

    Shimizu, J., Yamazaki, S., Takahashi, T., Ishida, Y. & Sakaguchi, S. Stimulation of CD25+CD4+ regulatory T cells through GITR breaks immunological self-tolerance. Nat. Immunol. 3, 135–142 (2002).

    CAS  Article  Google Scholar 

  13. 13

    Kohm, A.P., Williams, J.S. & Miller, S.D. Cutting edge: ligation of the glucocorticoid-induced TNF receptor enhances autoreactive CD4+ T cell activation and experimental autoimmune encephalomyelitis. J. Immunol. 172, 4686–4690 (2004).

    CAS  Article  Google Scholar 

  14. 14

    Stephens, G.L. et al. Engagement of glucocorticoid-induced TNFR family-related receptor on effector T cells by its ligand mediates resistance to suppression by CD4+CD25+ T cells. J. Immunol. 173, 5008–5020 (2004).

    CAS  Article  Google Scholar 

  15. 15

    Ronchetti, S. et al. GITR, a member of the TNF receptor superfamily, is costimulatory to mouse T lymphocyte subpopulations. Eur. J. Immunol. 34, 613–622 (2004).

    CAS  Article  Google Scholar 

  16. 16

    Ko, K. et al. Treatment of advanced tumors with agonistic anti-GITR mAb and its effects on tumor-infiltrating Foxp3+CD25+CD4+ regulatory T cells. J. Exp. Med. 202, 885–891 (2005).

    CAS  Article  Google Scholar 

  17. 17

    Seo, S.K. et al. 4–1BB-mediated immunotherapy of rheumatoid arthritis. Nat. Med. 10, 1088–1094 (2004).

    CAS  Article  Google Scholar 

  18. 18

    Asselin-Paturel, C. et al. Mouse type I IFN-producing cells are immature APCs with plasmacytoid morphology. Nat. Immunol. 2, 1144–1150 (2001).

    CAS  Article  Google Scholar 

  19. 19

    Colonna, M., Trinchieri, G. & Liu, Y.J. Plasmacytoid dendritic cells in immunity. Nat. Immunol. 5, 1219–1226 (2004).

    CAS  Article  Google Scholar 

  20. 20

    Ochando, J.C. et al. Alloantigen-presenting plasmacytoid dendritic cells mediate tolerance to vascularized grafts. Nat. Immunol. 7, 652–662 (2006).

    CAS  Article  Google Scholar 

  21. 21

    Mellor, A.L. et al. Cutting edge: induced indoleamine 2,3 dioxygenase expression in dendritic cell subsets suppresses T cell clonal expansion. J. Immunol. 171, 1652–1655 (2003).

    CAS  Article  Google Scholar 

  22. 22

    Fallarino, F. et al. Murine plasmacytoid dendritic cells initiate the immunosuppressive pathway of tryptophan catabolism in response to CD200 receptor engagement. J. Immunol. 173, 3748–3754 (2004).

    CAS  Article  Google Scholar 

  23. 23

    Fallarino, F. & Puccetti, P. Toll-like receptor 9-mediated induction of the immunosuppressive pathway of tryptophan catabolism. Eur. J. Immunol. 36, 8–11 (2006).

    CAS  Article  Google Scholar 

  24. 24

    Fallarino, F. et al. Ligand and cytokine dependence of the immunosuppressive pathway of tryptophan catabolism in plasmacytoid dendritic cells. Int. Immunol. 17, 1429–1438 (2005).

    CAS  Article  Google Scholar 

  25. 25

    Muller, A.J., DuHadaway, J.B., Donover, P.S., Sutanto-Ward, E. & Prendergast, G.C. Inhibition of indoleamine 2,3-dioxygenase, an immunoregulatory target of the cancer suppression gene Bin1, potentiates cancer chemotherapy. Nat. Med. 11, 312–319 (2005).

    CAS  Article  Google Scholar 

  26. 26

    Lawrence, T., Bebien, M., Liu, G.Y., Nizet, V. & Karin, M. IKKα limits macrophage NF-κB activation and contributes to the resolution of inflammation. Nature 434, 1138–1143 (2005).

    CAS  Article  Google Scholar 

  27. 27

    Bonizzi, G. & Karin, M. The two NF-κB activation pathways and their role in innate and adaptive immunity. Trends Immunol. 25, 280–288 (2004).

    CAS  Article  Google Scholar 

  28. 28

    Kinoshita, D. et al. Essential role of IκB kinase α in thymic organogenesis required for the establishment of self-tolerance. J. Immunol. 176, 3995–4002 (2006).

    CAS  Article  Google Scholar 

  29. 29

    Chen, X., Murakami, T., Oppenheim, J.J. & Howard, O.M. Differential response of murine CD4+CD25+ and CD4+CD25 T cells to dexamethasone-induced cell death. Eur. J. Immunol. 34, 859–869 (2004).

    CAS  Article  Google Scholar 

  30. 30

    Romani, L. Immunity to fungal infections. Nat. Rev. Immunol. 4, 1–23 (2004).

    Article  Google Scholar 

  31. 31

    Judson, M.A. & Stevens, D.A. Current pharmacotherapy of allergic bronchopulmonary aspergillosis. Expert Opin. Pharmacother. 2, 1065–1071 (2001).

    CAS  Article  Google Scholar 

  32. 32

    Montagnoli, C. et al. Immunity and tolerance to Aspergillus involve functionally distinct regulatory T cells and tryptophan catabolism. J. Immunol. 176, 1712–1723 (2006).

    CAS  Article  Google Scholar 

  33. 33

    Romani, L. et al. Thymosin α1 activates dendritic cell tryptophan catabolism and establishes a regulatory environment for balance of inflammation and tolerance. Blood 108, 2265–2274 (2006).

    CAS  Article  Google Scholar 

  34. 34

    Romani, L. & Puccetti, P. Protective tolerance to fungi: the role of IL-10 and tryptophan catabolism. Trends Microbiol. 14, 183–189 (2006).

    CAS  Article  Google Scholar 

  35. 35

    Bluestone, J.A. & Tang, Q. How do CD4+CD25+ regulatory T cells control autoimmunity? Curr. Opin. Immunol. 17, 638–642 (2005).

    CAS  Article  Google Scholar 

  36. 36

    Grohmann, U., Fallarino, F. & Puccetti, P. Tolerance, DCs and tryptophan: much ado about IDO. Trends Immunol. 24, 242–248 (2003).

    CAS  Article  Google Scholar 

  37. 37

    Orabona, C. et al. Towards the identification of a tolerogenic signature in IDO-competent dendritic cells. Blood 107, 2846–2854 (2006).

    CAS  Article  Google Scholar 

  38. 38

    Gurtner, G.J., Newberry, R.D., Schloemann, S.R., McDonald, K.G. & Stenson, W.F. Inhibition of indoleamine 2,3-dioxygenase augments trinitrobenzene sulfonic acid colitis in mice. Gastroenterology 125, 1762–1773 (2003).

    CAS  Article  Google Scholar 

  39. 39

    Hayashi, T. et al. Inhibition of experimental asthma by indoleamine 2,3-dioxygenase. J. Clin. Invest. 114, 270–279 (2004).

    CAS  Article  Google Scholar 

  40. 40

    de Heer, H.J. et al. Essential role of lung plasmacytoid dendritic cells in preventing asthmatic reactions to harmless inhaled antigen. J. Exp. Med. 200, 89–98 (2004).

    CAS  Article  Google Scholar 

  41. 41

    Mellor, A.L. et al. Cutting edge: CpG oligonucleotides induce splenic CD19+ dendritic cells to acquire potent indoleamine 2,3-dioxygenase-dependent T cell regulatory functions via IFN type 1 signaling. J. Immunol. 175, 5601–5605 (2005).

    CAS  Article  Google Scholar 

  42. 42

    Wingender, G. et al. Systemic application of CpG-rich DNA suppresses adaptive T cell immunity via induction of IDO. Eur. J. Immunol. 36, 12–20 (2006).

    CAS  Article  Google Scholar 

  43. 43

    Yang, C.H., Murti, A. & Pfeffer, L.M. Interferon induces NF-κB-inducing kinase/tumor necrosis factor receptor-associated factor-dependent NF-κB activation to promote cell survival. J. Biol. Chem. 280, 31530–31536 (2005).

    CAS  Article  Google Scholar 

  44. 44

    Bonizzi, G. et al. Activation of IKKα target genes depends on recognition of specific κB binding sites by RelB:p52 dimers. EMBO J. 23, 4202–4210 (2004).

    CAS  Article  Google Scholar 

  45. 45

    Gilliet, M. & Liu, Y.J. Generation of human CD8 T regulatory cells by CD40 ligand-activated plasmacytoid dendritic cells. J. Exp. Med. 195, 695–704 (2002).

    CAS  Article  Google Scholar 

  46. 46

    Ito, T. et al. Plasmacytoid dendritic cells prime IL-10-producing T regulatory cells by inducible costimulator ligand. J. Exp. Med. 204, 105–115 (2007).

    CAS  Article  Google Scholar 

  47. 47

    Hwu, P. et al. Indoleamine 2,3-dioxygenase production by human dendritic cells results in the inhibition of T cell proliferation. J. Immunol. 164, 3596–3599 (2000).

    CAS  Article  Google Scholar 

  48. 48

    Fallarino, F. et al. The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor ζ-chain and induce a regulatory phenotype in naive T cells. J. Immunol. 176, 6752–6761 (2006).

    CAS  Article  Google Scholar 

  49. 49

    Karagiannidis, C. et al. Glucocorticoids upregulate FOXP3 expression and regulatory T cells in asthma. J. Allergy Clin. Immunol. 114, 1425–1433 (2004).

    CAS  Article  Google Scholar 

  50. 50

    Vacca, C. et al. CD40 ligation prevents onset of tolerogenic properties in human dendritic cells treated with CTLA-4-Ig. Microbes Infect. 7, 1040–1048 (2005).

    CAS  Article  Google Scholar 

Download references


We thank P. Mosci for maintaining the mutant strains of mice and performing histopathology; and G. Andrielli for digital art and image editing. Supported by the Italian Association for Cancer Research.

Author information



Corresponding author

Correspondence to Paolo Puccetti.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Fig. 1

GITR-Ig mediates processing of p100 to generate p52. (PDF 171 kb)

Supplementary Fig. 2

Kinetic immunoblot analysis of NIK, IKKα and IKKβ expression in pDCs treated with specific siRNAs (+) in one experiment representative of three. (PDF 196 kb)

Supplementary Fig. 3

NIK and IKKα are required for IFN-α induction by GITR-Ig. (PDF 21 kb)

Supplementary Fig. 4

Dexamethasone in vivo up-regulates GITR and GITRL. (PDF 83 kb)

Supplementary Fig. 5

Cytokine production in vitro in response to a range of GITR-Ig concentrations by pDCs from mice treated or not with dexamethasone (dex). (PDF 23 kb)

Supplementary Fig. 6

GITR-Ig activates noncanonical NF-κB signaling in Ifnar−/− mice. (PDF 242 kb)

Supplementary Methods (PDF 139 kb)

Supplementary Note (PDF 197 kb)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Grohmann, U., Volpi, C., Fallarino, F. et al. Reverse signaling through GITR ligand enables dexamethasone to activate IDO in allergy. Nat Med 13, 579–586 (2007).

Download citation

Further reading


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing