Interferon-γ and IL-27 positively regulate type 1 regulatory T cell development during adaptive tolerance

Abstract

Strong T cell receptor (TCR) and interleukin (IL)-27 signaling influence type 1 regulatory (Tr1) T cell development, but whether other signals determine their differentiation is unclear. Utilizing Tg4 TCR transgenic mice, we established a model for rapid Tr1 cell induction. A single high dose of [4Y]-MBP peptide drove the differentiation of Il10 ⁺ T cells with Tr1 cell mRNA and protein signatures. Kinetic transcriptional and phenotypic analyses revealed that the Tr1 cell module was transient and preceded by Ifng transcription in other CD4⁺ T cells. Changes in Tr1 cell frequency correlated with altered macrophage activation, while neutralization of interferon (IFN)γ reduced Tr1 cell frequency and the TCR signal strength markers Nur77, inducible T cell costimulator (ICOS), and OX40. Antibody depletion experiments inferred that the relevant source of IFNγ was not natural killer (NK) cell derived. Additionally, blocking IL-27 in combination with IFNγ neutralization additively reduced Tr1 cell frequency in vivo. These findings reveal that IFNγ has a non-redundant role in augmenting Tr1 cell differentiation in vivo.

Keywords: Cell biology; Immunology; Stem cell research.

Graphical abstract

Figure 1

Kinetics of de novo Il10 transcription in response to strong tolerogenic TCR signaling Tg4 Nr4a3-Tocky Il10-EGFP mice were immunized with 0.8, 8, or 80 μg [4Y]-MBP in 200 μL PBS subcutaneously (s.c.), and spleens were harvested at 24 h. Representative flow plots (A) showing CD4⁺ TCRvβ8.1/8.2⁺ according to Nr4a3-Timer expression above and the Il10-EGFP⁺ of Nr4a3-Timer⁺ and Nr4a3-Timer⁻ below. Summary of Il10-EGFP frequency (B), LAG3 MFI (C), and TIGIT MFI (D) in Nr4a3-Timer⁺ (orange) or Nr4a3-Timer^- (red). Summary of LAG3 (E) and TIGIT MFI (F) in Il10-EGFP⁺Nr4a3-Timer⁺. Tg4 Nr4a3-Tocky Il10-EGFP Ifng-YFP mice were administered 80 μg [4Y]-MBP in 200 μL PBS s.c., and spleens were harvested at 8, 12, or 16 h. Representative flow plots (G) showing CD4⁺ TCRvβ8.1/8.2⁺Nr4a3-Timer expression above and below Il10-EGFP against Ifng-YFP of Nr4a3-Timer⁺ below. Summary of Il10-EGFP⁺Ifng-YFP⁻ (H) and Ifng-YFP⁺Il10-EGFP⁻ (I) frequency, LAG3 (J), TIGIT (K), PD-1 (L), CD25 (M), CD44 (N), and CD69 (O) MFI in Nr4a3-Timer^±. (B–F and H–O) bars represent mean ± SEM. Statistical analysis by two-way ANOVA with Tukey’s multiple comparisons test (B–D and H–O) or Kruskal-Wallis test with Dunn’s multiple comparisons. n = 3 per treatment or time.

Figure 2

Rapidly induced Il10-expressing T cells are Tr1-like cells and have a transcriptionally delayed program (A) Re-analysis of GEO: GSE165817 (Elliot et al.¹⁹) where Tg4 Nr4a3-Tocky Il10-EGFP mice were administered with 0 (control), 8, or 80 μg [4Y]-MBP in 200 μL PBS s.c. and spleens harvested at 4, 12, and 24 h for bulk RNA isolation and transcriptome analysis. Heatmap of markers associated with early T cell activation, Th1 phenotype or Tr1-like cells arising after expression of Th1 cell hallmarks. (B) Gene set enrichment analysis (GSEA) using Tr1 genes defined in Chen et al. comparing 12 h gene expression in 80 μg vs. 0.8 μg immunized mice. (C) Heatmap of rlog-normalized data for GSEA core-enrichment genes for weak (0.8 μg) versus strong (80 μg) conditions. Tg4 Nr4a3-Tocky Il10-EGFP mice were administered 4 mg/kg [4Y]-MBP in 200 μL PBS s.c., and 24 h later CD4⁺Il10-EGFP+ and CD4⁺Il10-EGFP⁻ T cells were sorted for analysis by mRNA-seq. (D) Transcriptome analysis of CD4⁺Nr4a3-Timer⁺Il10-EGFP^± as a heatmap of curated Tr1 differentially expressed genes (DEGs) between conditions at 24 h. (E–G) Compared temporal expression rlog transcripts from GEO: GSE165817 (Elliot et al.¹⁹) for Ifng against Il10 (E), Ifng against Lag3 (F), and Lag3 against Il10 (G) from Tg4 Nr4a3-Timer Il10-EGFP mice administered with 80 μg [4Y]-MBP in 200 μL PBS s.c. and mRNA transcripts depicted at 4, 12, and 24 h.

Figure 3

Interferon-gamma positively regulates Tr1 cell differentiation Tg4 Nr4a3-Tocky Il10-EGFP Ifng-YFP mice were immunized with 4 mg/kg [4Y]-MBP in PBS s.c. and 1 mg αIFNγ or IgG1 isotype in 200 μL PBS i.p., and spleens were harvested 24 h later. Representative flow plots at 24 h (A) showing CD4⁺ TCRvβ8.1/8.2⁺ according to Nr4a3-Timer expression above and the Il10-EGFP against Ifng-YFP of Nr4a3-Timer⁺ below. Summary of Il10-EGFP (B) and Ifng-YFP (C) frequency in Nr4a3-Timer⁺ and CD69 (D), TIGIT (E), LAG3 (F), ICOS (G), and GITR (H) MFI. Tg4 Nr4a3-Tocky Il10-EGFP Ifng-YFP mice were immunized with 4 mg/kg [4Y]-MBP in PBS s.c. and 1 mg αIFNγ or IgG1 isotype in 200 μL PBS i.p., and spleens were harvested at 12 h (I) showing CD11b against MHC Class II to the left, and from CD11b⁺, F4/80 against MHC Class II to the right. Summary of F4/80⁺ MHCII⁺ (“Inflammatory” myeloid) frequency (J) CD40 (K), PD-L1 (L), CD86 (M), and MHC Class II (N) MFI. Linear regression (Pearson’s correlation p value and r value in red) of CD4⁺Nr4a3-Timer⁺Il10-EGFP⁺ frequency against CD40 (O), CD86 (P), MHC Class II (Q), and PD-L1 (R) MFI in the inflammatory myeloid compartment. (B–H and I–N) Bars represent median with interquartile range. Statistical analysis by Mann-Whitney U test. n = 8 (A–I) and n = 7 (J–R) per treatment.

Figure 4

Anti-IFNγ treatment reduces TCR signal strength in vivo Tg4 Nur77-Tempo Il10-EGFP mice received 1 mg αIFNγ or IgG1 isotype in 200 μL PBS i.p before immunization with 4 mg/kg [4Y]-MBP in PBS s.c., and spleens were harvested at 24 h. (A and B) (A) Nur77-Blue versus Nur77-Red or (B) Il10-EGFP vs. CD4 in CD4⁺ T cells at 24 h. (C–I) (C) Summary of % Nur77-Blue⁺Red⁺, (D) Nur77-Blue MFI in the two treatment, or % Il10-EGFP (E) in CD4⁺ T cells. ICOS (F) or OX40 (G) MFI in CD4⁺ T cells. Correlation of Nur77-Blue MFI versus ICOS MFI (H) or OX40 MFI (I) in CD4⁺ T cells using linear regression (Pearson’s correlation p value and r² values are shown). Isotype (n = 7), αIFNγ (n = 8). Bars represent median+/−interquartile range (IQR). Statistical analysis by Mann-Whitney U test.

Figure 5

IFNγ and IL-27 additively regulate Tr1 cell development Tg4 Nr4a3-Tocky Il10-EGFP Ifng-YFP mice were administered 4 mg/kg [4Y]-MBP in PBS s.c. and 1 mg αIFNγ or IgG1 isotype in 200 μL PBS i.p., and spleens were harvested at 24 h. Representative flow plots showing Il10-EGFP against Ifng-YFP above and Ifng-YFP⁺-derived TCRvβ8.1/8.2 against NK1.1 below (A) and TCRvβ8.1/8.2 against NK1.1 above and NK1.1⁺-derived Il10-EGFP against Ifng-YFP below (B). Summary of Ifng-YFP⁺ frequency (C) from (A) and Ifng-YFP⁺ frequency from NK1.1⁺ (D) from (B). Linear regression (Pearson’s correlation) of CD4⁺Nr4a3-Timer⁺Il10-EGFP⁺ against NK1.1⁺ TCRvβ8.1/8.2⁻ (E). Tg4 Nr4a3-Tocky Il10-EGFP mice were given 200 μg αNK1.1 or IgG2a isotype in 200 μL PBS i.p. and 48 h later were immunized with 4 mg/kg [4Y]-MBP in PBS s.c., and spleens were harvested 24 h later. Representative spectral cytometry plots showing NK1.1 against FSC-A (F). Summary of NK1.1⁺ (G) from (F). Representative spectral cytometry plots at 24 h showing Il10-EGFP against CD4 (H). Summary of Il10-EGFP⁺ frequency (I) and ICOS (J) and OX40 (K) MFI of CD4⁺Nr4a3-Timer⁺ T cells. Tg4 Nr4a3-Tocky Il10-EGFP mice were administered 4 mg/kg [4Y]-MBP in PBS s.c. and 0.5 mg αIFNγ, αIL-27, a combination of both, or a mixed IgG1 and IgG2a isotype in 200 μL PBS i.p., and spleens were harvested 24 h later. Representative spectral cytometry plots at 24 h showing Il10-EGFP against CD4 in Nr4a3-Timer⁺ (L). Summary of Il10-EGFP⁺ frequency within the CD4⁺Nr4a3-Timer⁺ T cells. (M) Analysis for additive effect of treatment by factoring αIFNγ and αIL-27 as two variables (N).(C, D, G, I–K, M, and N) Bars represent median with interquartile range. Statistical analysis by Mann-Whitney U test (C, D, and I–K), Kruskal-Wallis test with Dunn’s multiple comparisons (M). n = 4 (A–E), n = 8 (F–K), n = 15 (Combo), n = 16 (isotype), n = 17 (αIFNγ), and n = 17 (αIL-27) (L–N) per treatment.