Forced IDO1 expression in dendritic cells restores immunoregulatory signalling in autoimmune diabetes

Abstract

Indoleamine 2,3-dioxygenase (IDO1), a tryptophan catabolizing enzyme, is recognized as an authentic regulator of immunity in several physiopathologic conditions. We have recently demonstrated that IDO1 does not merely degrade tryptophan and produce immunoregulatory kynurenines, but it also acts as a signal-transducing molecule, independently of its enzymic function. IDO1 signalling activity is triggered in plasmacytoid dendritic cells (pDCs) by transforming growth factor-β (TGF-β), an event that requires the non-canonical NF-κB pathway and induces long-lasting IDO1 expression and autocrine TGF-β production in a positive feedback loop, thus sustaining a stably regulatory phenotype in pDCs. IDO1 expression and catalytic function are defective in pDCs from non-obese diabetic (NOD) mice, a prototypic model of autoimmune diabetes. In the present study, we found that TGF-β failed to activate IDO1 signalling function as well as up-regulate IDO1 expression in NOD pDCs. Moreover, TGF-β-treated pDCs failed to exert immunosuppressive properties in vivo. Nevertheless, transfection of NOD pDCs with Ido1 prior to TGF-β treatment resulted in activation of the Ido1 promoter and induction of non-canonical NF-κB and TGF-β, as well as decreased production of the pro-inflammatory cytokines, interleukin 6 (IL-6) and tumour necrosis factor-α (TNF-α). Overexpression of IDO1 in TGF-β-treated NOD pDCs also resulted in pDC ability to suppress the in vivo presentation of a pancreatic β-cell auto-antigen. Thus, our data suggest that a correction of IDO1 expression may restore its dual function and thus represent a proper therapeutic manoeuvre in this autoimmune setting.

Keywords: IDO1; autoimmune diabetes; immune regulation; non-canonical NF-κB; non-obese diabetic (NOD) mice; plasmacytoid dendritic cells; tryptophan catabolism.

Figure 1

Transforming growth factor-β (TGF-β) fails to confer suppressive properties and induce IDO1 expression in pDCs from NOD mice. In vivo suppressive activity of TGF-β–conditioned pDCs from C57BL/6, BALB/c and NOD mice (A). Purified splenic cDCs and pDCs were pulsed for 2 hr with the HY (C57BL/6 groups) or IGRP (BALB/c and NOD) peptide and transferred into syngeneic recipient mice to be assayed for skin reactivity to the eliciting peptide. The cDC fraction was used in combination with 5% pDCs, pre-conditioned by TGF-β or medium alone, that were left untransfected, transfected with siRNA containing a scrambled sequence (negative control, NC) or targeting Ido1 (only in C57BL/6 cells). Analysis of skin reactivity of recipient mice to the eliciting peptide at 15 d is presented as change in footpad weight. C57BL/6 and NOD pDCs were incubated with medium alone (control) or TGF-β for 16 hrs and Ido1 mRNA was either qualitatively (B) or quantitatively (C) analyzed by conventional or real-time PCR, respectively, by using Gapdh normalization. In (C), data are presented as normalized transcript expression in the samples relative to normalized transcript expression in the control culture (that is, fold change = 1, dotted line). (D) C57BL/6 and NOD pDCs were treated with TGF-β or medium alone for 24 hrs and IDO1 expression was assessed by immunoblot analysis, by using an IDO1-specific antibody. (E) IDO1 catalytic activity in C57BL/6 and NOD pDCs left untreated or treated for 24 hrs with TGF-β was assessed as l-kynurenine in culture supernatants. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). Results are representative of three (A,B and C), five (D) and four (E) experiments (mean ± SD in A,B and D).

Figure 2

Transforming growth factor β (TGF-β) confers suppressive properties in NOD pDCs overexpressing IDO1. (A) NOD pDCs were transfected with mRNA coding for wtIDO1, IDO1.Y₁₁₅FY₂₅₃F or IDO1.H₃₅₀A and, after 24 hrs, Ido1 mRNA was quantified by real-time PCR by using Gapdh normalization. Irrelevant mRNA was used as a control (control). Data are presented as normalized transcript expression in the samples relative to normalized transcript expression in the untransfected cell culture (that is, fold change = 1, dotted line). (B) IDO1 catalytic activity in NOD pDCs transfected as in (A) was assessed as l-kynurenine in culture supernatants. (C) NOD pDCs were transfected as in (A) and treated with TGF-β prior to pulsing with IGRP and administration into recipient mice as in Figure1A. Analysis of skin reactivity of recipient mice to the eliciting peptide at 15 d is presented as change in footpad weight. (D) NOD pDCs transfected with wtIDO1 were treated with TGF-β in the presence or absence (no inhibitor) of 1-MT, the standard IDO1 inhibitor, PP2, a Fyn inhibitor or PP3 (negative control of PP2). After peptide pulsing with IGRP, cells were administered into recipient mice as in (A and C) and analysis of skin reactivity was performed at 15 d post-sensitization. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). Results are representative of three (A,C,D) and four (B) experiments (mean ± SD).

Figure 3

IDO1 transfection combined with transforming growth factor-β (TGF-β) treatment modulates cytokine production by NOD pDCs. (A) Cytokine analysis in culture supernatants. NOD pDCs were transfected with irrelevant mRNA (control) or the construct coding for wtIDO1 prior to treatment with TGF-β (at 24 hrs post-transfection). Untransfected splenic pDCs, either untreated (vehicle) or treated with TGF-β, from NOD and C57BL/6 mice were analysed for comparison. Supernatants were harvested at 24 hrs of TGF-β incubation and assayed for cytokine contents by ELISA. Results are means ± SD of four experiments. (B) Cytofluorometric analysis of LAP TGF-β expression. Cells from the same groups as in (A) were co-stained with B220– (a pDC marker) and LAP TGF-β–specific antibodies and analysed by cytofluorometric analysis. Results represent percentages of B220⁺LAP TGF-β⁺ cells. Upper panel, dot plots of most representative groups from one experiment by using NOD pDCs. Lower panel, means ± SD of three experiments. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test).

Figure 4

Transforming growth factor-β (TGF-β) activates the non-canonical NF-κB pathway and Ido1 promoter in NOD pDCs overexpressing IDO1. (A) Time course of activation of the Ido1 promoter in pDCs (same groups as in Fig.3) transfected with a firefly luciferase construct of the Ido1 promoter and incubated for 24–48 hrs with TGF-β; results are normalized to the activity of a cotransfected constitutive reporter and are presented relative to those in cells not treated with TGF-β (dashed line, onefold). (B) ELISA of the activation of p65, p52 and RelB in nuclear extracts of pDCs not treated or treated with TGF-β for 30 min. Results are presented as absorbance at 450 nm (A₄₅₀). (C) Immunoblot analysis of the relative expression of p100 versus p52. Whole cell lysates of C57BL/6 and NOD pDCs stimulated with TGF-β for 30 min. were immunoblotted with anti-p100 and anti-p100/p52 antibodies. β-tubulin was used as loading control. One representive experiment is shown. (D) Ratios of p52:p100 were measured by densitometric quantification of the bands detected in three experiments (means ± SD) one of which is shown in (C). Results are representative of three (A) and two (B) experiments (means ± SD). *P < 0.01 and **P < 0.01, cytokine-treated versus untreated (Student's t-test).