IFN-α and CD46 stimulation are associated with active lupus and skew natural T regulatory cell differentiation to type 1 regulatory T (Tr1) cells

Abstract

Immune suppressive activities exerted by regulatory T-cell subsets have several specific functions, including self-tolerance and regulation of adaptive immune reactions, and their dysfunction can lead to autoimmune diseases and contribute to AIDS and cancer. Two functionally distinct regulatory T-cell subsets are currently identified in peripheral tissues: thymus-developed natural T regulatory cells (nTregs) controlling self-tolerance and antiinflammatory IL-10-secreting type 1 regulatory T cells (Tr1) derived from Ag-stimulated T cells, which regulate inflammation-dependent adaptive immunity and minimize immunopathology. We establish herein that cell contact-mediated nTreg regulatory function is inhibited by inflammation, especially in the presence of the complement C3b receptor (CD46). Instead, as with other T-cell subsets, the latter inflammatory conditions of stimulation skew nTreg differentiation to Tr1 cells secreting IL-10, an effect potentiated by IFN-α. The clinical relevance of these findings was verified in a study of 152 lupus patients, in which we showed that lupus nTreg dysfunction is not due to intrinsic defects but is rather induced by C3b stimulation of CD46 and IFN-α and that these immune components of inflammation are directly associated with active lupus. These results provide a rationale for using anti-IFN-α Ab immunotherapy in lupus patients.

Fig. 1.

Context dependency of nTreg physiologic characteristics. nTregs (4 × 10⁴ per well) were stimulated with platebound anti-CD3 mAb (pbαCD3) (0.2 μg/mL) in presence of irradiated ΔCD3-feeder (4 × 10⁴ per well) and where indicated, with soluble anti-CD28 mAb (sαCD28) (4 μg/mL) or IL-2 (100 IU/mL). CFSE-labeled nTregs (nTregs^CFSE) were cultured without (Left) or with (Right) Tconv, in response to various stimulations ([1]–[3]). (A) nTregs^CFSE cell activation status, evaluated by the median fluorescence intensity of CD25 expression: FACS histograms. (B) nTregs^CFSE cell proliferation, assessed by the CFSE dilution assay: FACS histograms. Data are representative of four independent experiments. (C) nTreg suppressive assay. CFSE-labeled Tconv (Tconv^CFSE) were cocultured with nTregs at different ratios under the indicated stimulations ([1]–[3]). Proliferation of Tconv^CFSE was evaluated by the CFSE dilution assay: (C1) FACS histogram and (C2) mean % ± SD of Tconv^CFSElow (n = 4). (D) Evidence for cell contact-mediated nTreg suppression. Proliferation of Tconv^CFSE cocultured with nTregs at the indicated ratios either (D1) with 96-transwell inserts or not (control) (n = 3) or (D2) in presence of anti–IL-10/anti–IL-10 receptor Abs or isotypic-Abs (control) (n = 3). (E) IL-2 effect on nTreg suppressive function. (E1) Inhibition of the cell contact suppressive activity of nTregs assessed in CD3-stimulated nTreg–Tconv^CFSE cocultures in presence of increasing doses of sαCD28 mAb: the inhibition is dependent on CD28 stimulation magnitude and abrogated by anti–IL-2 Abs (n = 3). (E2) Cell contact suppressive activity of nTregs in CD3-stimulated cocultures is inhibited by exogenous IL-2 in a dose-dependent manner (n = 3). (E3) IL-2 levels quantified by ELISA in the 4-d culture supernatant of stimulated Tconv (4 × 10⁴ per well) (mean ± SD) (n = 3).

Fig. 2.

Effect of inflammatory components on nTreg physiology. (A) Effect of CD46 stimulation on nTregs. CD3-stimulated nTregs^CFSE were cultured in presence or absence of sαCD28 mAb without or with Tconv as described in Fig.1 and additional stimulatory signals provided by plate-bound anti-CD46 mAb (pbαCD46) (10 μg/mL). (A1) nTreg^CFSE cell activation and (A2) nTreg^CFSE cell proliferation: mean % ± SD of CD25 median fluorescence intensity (MFI) and mean % ± SD of nTregs^CFSElow, respectively (n = 4). (A3) nTreg suppressive assay: mean % ± SD of Tconv^CFSElow (n = 4). (B) IFN-α effects on stimulated nTregs compared with those on Tconv. nTreg^CFSE (4 × 10⁴ per well) were cocultured with Tconv at a 1:1 ratio in the presence of ΔCD3-feeder (4 × 10⁴ per well), pbαCD3 mAb (0.5 μg/mL), sαCD28 mAb (1 μg/mL), and increasing concentrations of IFN-α (0.1–100 ng/mL) for 4 d. In a parallel experiment, Tconv^CFSE were cultured alone under the same stimulation conditions as those described above. (B1) T-cell activation status (CD25 MFI) on both CFSE-labeled T-cell subsets, measured by flow cytometry, and (B2) T-cell proliferation measured by CFSE dilution assay: IFN-α dose–effect inhibition (mean % ± SD) (n = 3). (B3) Apoptosis of 6 d-cultured nTregs and Tconv (2 × 10⁵ per well) stimulated with pbαCD3 mAb (0.5 μg/mL) in presence of sαCD28 mAb (4 μg/mL), IL-2 (100 IU/mL), without or with IFN-α (10 ng/mL) was evaluated by 7-amino-actinomycin D staining (35): mean % (apoptotic cells + apoptotic bodies/debris) ± SD of four experiments. (C) IL-10 production by Tr1 differentiated nTregs compared with that of Tconv. nTregs and Tconv (2 × 10⁵ per well) were stimulated with pbαCD3 mAb (0.5 μg/mL) in presence of sαCD28 mAb (4 μg/mL), IL-2 (100 IU/mL), without or with IFN-α (10 ng/mL). Where indicated, additional costimulatory signals were provided by pbαCD46 mAb (10 μg/mL). (C1) IL-10 mRNA levels measured by RT-PCR in the 36 h-cultured T-cell subsets (mean ± SD) (n = 3). (C2) IL-10 levels quantified by ELISA in the 4 d-culture supernatant of nTregs and Tconv stimulated as indicated (mean ± SD) (n = 3).

Fig. 3.

nTreg and Tconv differentiation toward Tr1 cells. (A1) Experimental scheme for assessing Tr1 cytokine-dependent suppressive function, using 96-transwell plates and appropriate cell number. In the bottom chamber, nTregs or Tconv (3 × 10⁵ per well) were stimulated with pbαCD3 mAb (0.5 μg/mL) in presence of ΔCD3-feeder. Where indicated, additional costimulatory signals were provided by sαCD28 mAb (4 μg/mL), pbαCD46 mAb (10 μg/mL), or IL-2 (20 IU/mL). Sixteen hours later, cells were washed and were cultured without (A2) or with (A3) CD3/CD28-prestimulated Tconv^CFSE-responder cells (3 × 10⁴ per well) in presence of ΔCD3-feeder (upper chamber: insert) and fresh medium with IL-2 where indicated. (A2) IL-10 release by bottom cells stimulated as indicated, assessed by ELISA (n = 3). (A3) Inhibition of prestimulated Tconv^CFSE-responder cell proliferation by soluble factors released by bottom cells, as evaluated by the CFSE dilution assay: FACS histograms (Left); mean suppression % ± SD (n = 3) (Right). (B) Reversion to the original function of Tr1-differentiated nTregs. (B1) Purification of IL-10–secreting cells stimulated with CD3/CD28/+IL2±CD46 using the IL-10 secretion assay from Miltenyi-Biotec: FACS plots of pre- (Left) and post- (Right) sort. (B2) nTreg function assessed under steady state in anti-CD3 Ab-coated round-bottom wells (and not transwells) harboring cocultures of Tconv^CFSE-responders and either fresh nTregs (control) or 3 d-resting IL-10–secreting nTregs (nTreg^IL-10+) purified from CD3/CD28/+IL-2- ([1]) or CD3/CD28/CD46/+IL-2-stimulated nTregs ([2]): mean % suppression ± SD of Tconv^CFSElow proliferation (n = 3).

Fig. 4.

Lupus patient nTreg physiologic characteristics. (A) CD4 T cells and nTreg subset in HD, inactive (IL), and active (AL) lupus patients. (A1) Median % of CD4 T cells in PBMCs. (A2) Median % of nTreg (CD4⁺CD127⁻CD25^bright) within gated CD4 T cells in HD, IL, and AL patients. (B) Suppressive function of HD, IL, and AL nTregs evaluated in cocultures stimulated (B1) by anti-CD3 mAb and (B2) by anti-CD3/CD28/CD46 mAbs. nTreg suppressive activity at 1:1 nTreg:Tconv^CFSE ratio in different groups was evaluated by the CFSE dilution assay. (B1) Median % suppression ± SD; (B2) Mean % suppression ± SD. (C) IFN-α effects on CD3/CD28/+IL-2±CD46-stimulated lupus nTregs. (C1) IL-10 levels were quantified by ELISA in the 4 d-culture supernatant; mean % ± SD of three experiments. (C2 and C3) Apoptosis of 6 d-cultured nTregs stimulated as indicated was evaluated by 7-amino-actinomycin D (7-AAD) staining: (C2) FACS plots representation (apoptotic cells are FSC^high/7-AAD⁺ and apoptotic bodies/debris) (FSC^low); (C3) mean % (apoptotic cells + apoptotic bodies/debris) ± SD of three experiments. *P < 0.05; **P < 0.005; ***P < 0.001; ns, P > 0.05.

Fig. 5.

Lupus activity dependency on microenvironmental inflammatory factors. (A) Correlation between lupus activity (SLEDAI score) and serum levels of (A1) complement C3, (A2) IFN-α, (A3) (C3+IFN-α), and (A4) IL-10. (B) Comparison of median serum levels of (B1) C3, (B2) IFN-α, (B3) (C3+IFN-α) score, and (B4) IL-10 in different groups of subjects inactive (IL), midly active (mAL), highly active (hAL) patients, and HD (normal HD serum C3 = 0.9–1.5 g/L). *P < 0.05; **P < 0.005; ***P < 0.001; ns, P > 0.05.