Inflammation-induced Id2 promotes plasticity in regulatory T cells

Abstract

T_H17 cells originating from regulatory T (T_reg) cells upon loss of the T_reg-specific transcription factor Foxp3 accumulate in sites of inflammation and aggravate autoimmune diseases. Whether an active mechanism drives the generation of these pathogenic 'ex-Foxp3 T_H17' cells, remains unclear. Here we show that pro-inflammatory cytokines enhance the expression of transcription regulator Id2, which mediates cellular plasticity of T_reg into ex-Foxp3 T_H17 cells. Expression of Id2 in in vitro differentiated iT_reg cells reduces the expression of Foxp3 by sequestration of the transcription activator E2A, leading to the induction of T_H17-related cytokines. T_reg-specific ectopic expression of Id2 in mice significantly reduces the T_reg compartment and causes immune dysregulation. Cellular fate-mapping experiments reveal enhanced T_reg plasticity compared to wild-type, resulting in exacerbated experimental autoimmune encephalomyelitis pathogenesis or enhanced anti-tumor immunity. Our findings suggest that controlling Id2 expression may provide a novel approach for effective T_reg cell immunotherapies for both autoimmunity and cancer.

Fig. 1

Id2 expression is enhanced during a conversion to T_H17 phenotype from iT_reg cells. a In vitro polarization scheme of T_H0, T_H17, iT_reg and ex-Foxp3 T_H17 cells. b RT-qPCR analysis of Id2 mRNA in in vitro generated T_H0, T_H17, iT_reg and ex-Foxp3 T_H17 cells; results are presented relative to Hprt expression (n = 3) c, d Western blot analysis to determine the kinetics of indicated proteins during T_H17 (c, top panel) and iT_reg (d, top panel) cell differentiation and maintenance conditions. β-actin is used as loading control. RT-qPCR analysis of Il17a, Il17f (c, lower panel) and Il10 (d, lower panel) mRNA during each T_H17 and iT_reg cell differentiation conditions; results are presented relative to Hprt expression (n = 3). e Sorted CD4⁺ naive T cells were activated in vitro under iT_reg cell differentiation condition for 3 days. After 3 days, iT_reg cells were re-stimulated in vitro in ex-Foxp3 T_H17 polarizing conditions for additional 3 days. Western blot analyses for the indicated proteins were performed after harvesting the cells at indicated time points (top panel). RT-qPCR analysis of Il10 and Il17a mRNA under similar experimental conditions is shown in the lower panel; results are presented relative to Hprt expression (n = 3). *P < 0.05 (Student’s t-test). Data are representative two independent experiments (error bars, s.d.)

Fig. 2

Overexpression of Id2 in vitro fails to stabilize iT_reg lineage commitment, and convert them into T_H17-like cells. a Naive CD4⁺ T cells were sorted from wild-type C57BL/6 (B6) mice, and transduced with control vector (Empty RV) or vector encoding Id2 cDNA (Id2 RV) under iT_reg differentiation condition. After 3 days cells were harvested and Id2 expression was measured by RT-qPCR (n = 3, per group) and Western blot analysis. b Comparison of mRNA expression for T-bet, Gata3, Rorγt and Foxp3 between Empty RV and Id2 RV transduced T cells after 3 days post spinfection under iT_reg differentiation condition (n = 3, per group). c Flow cytometry analysis of Foxp3 expression between Empty RV or Id2 RV transduced (GFP⁺) and non-transduced (GFP⁻) iT_reg cells (n = 3, per group). d Comparison of mRNA expression for genes encoding the indicated cytokines between Empty RV and Id2 RV transduced T cells after 3 days post spinfection under iT_reg differentiation condition (n = 3, per group). e Flow cytometry analysis of IL-17A, IL-17F and IL-22 from Empty RV or Id2 RV transduced CD4⁺GFP⁺ iT_reg cells (n = 3, per group). NS, not significant, *P < 0.05, **P < 0.005 (Student’s t-test). All data are representative two or three independent experiments (error bars, s.d.)

Fig. 3

Mice with T_reg-specific ectopic expression of Id2 displayed enhanced conversion to ex-Foxp3 T_H17 cells from T_reg cells after induction of EAE. a Schematic representation the mouse model employed. Converging arrows in the lower panel indicate genotyping primers. b Genotyping PCR to detect the presence of Id2-EmGFP transgene in Id2^EmGFPFoxp3^YFP−Cre mice. c Flow cytometry analysis of YFP and EmGFP expression in CD4⁺CD25⁺ T cells from spleen (SP) and peripheral lymph nodes (pLN) of Foxp3^YFP−Cre and Id2^EmGFPFoxp3^YFP−Cre mice (left panel). Id2 expression was assessed between non-T_reg and T_reg cells by intracellular staining (right panel). d Schematic of experimental EAE model. e Representative flow cytometry analysis of Id2 expression between tdTomato⁺YFP⁺ (T_reg) and tdTomato⁺YFP⁻(ex-T_reg) cells in CD4⁺ T cells from 6 to 8 week-old R26TFoxp3^YFP−Cre mice on day 24 after induction of EAE. f Mean clinical scores in mice after induction of EAE (R26TFoxp3^YFP−Cre; n = 12, Id2^EmGFPR26TFoxp3^YFP−Cre; n = 9). g Representative Spinal Cord (SC) sections and hematoxylin and eosin (H&E) staining from EAE induced mice at day 24 after immunization (scale bar, 100 μm). h Flow cytometry analysis and percentages of Foxp3 sufficient (YFP⁺) and deficient (YFP⁻) populations among CD4⁺tdTomato⁺ in SP and SC from R26TFoxp3^YFP−Cre and Id2^EmGFPR26TFoxp3^YFP−Cre mice at day 22–24 after induction of EAE. i Representative flow cytometry analysis and percentages of IFN-γ⁺, IL-17A⁺ and IFN-γ⁺IL-17A⁺ in CD4⁺tdTomato⁺ YFP⁻ or CD4⁺tdTomato⁺ YFP⁺ T cells in total SP and SC cells stimulated with phorbol myristate acetate (PMA) and ionomycin for 6 h. NS,  not significant, *P < 0.05, **P < 0.005, ***P < 0.001 (Student’s t-test). All data are representative of two independent experiments (error bars, s.d.)

Fig. 4

T_reg-cell-specific overexpression of Id2 results in systemic autoimmunity in older mice. a A representative picture of 12–16 week-old Foxp3^YFP−Cre and Id2^EmGFPFoxp3^YFP−Cre mice (left) and spleen (SP) and peripheral lymph nodes (pLN) derived from them (right). b Representative hematoxylin and eosin (H&E)-stained skin, lung, liver sections from 12–16 weeks old Foxp3^YFP−Cre and Id2^EmGFPFoxp3^YFP−Cre mice (scale bar, 100 μm). c Representative FACS plots and percentages of CD4⁺ and CD8⁺ T cells from SP and pLN of 12–16 weeks old Foxp3^YFP−Cre and Id2^EmGFPFoxp3^YFP−Cre mice. d, e Representative FACS plots and percentages of CD62L^hiCD44^lo (naive) and CD62L^loCD44^hi (effector/memory) cells among CD4⁺ and CD8⁺ T cells in SP and pLN 12–16 week-old mice. f Representative FACS plots and percentages of Foxp3⁺ T_reg cells in thymus, SP and pLN from 12–16 week-old mice. NS,  not significant, *P < 0.05, **P < 0.005 (Student’s t-test). All data are representative three independent experiments (error bars, s.d.)

Fig. 5

Id2 is induced by IL-1β and IL-6 mediated signaling and downstream transcription factors STAT3, IRF4 and BATF in T_reg cells. a Strategy for sorting T_Naive, iT_reg, and ex-Foxp3 T_H17 cells from Foxp3^Thy1.1 reporter mice (left). Flow cytometry analysis of IL-1r1 and IL-6rα among the T_Naive, iT_reg, and ex-Foxp3 T_H17 cells (right). b, c FACS-sorted iT_reg cells were re-stimulated with cytokines as indicated and analyzed for Id2 and Foxp3 mRNA expression by RT-qPCR (b) as well as for Id2 protein levels by flow cytometry (c). d Cartoon depicting highly putative STAT3, IRF4, and BATF binding motifs upstream of Id2 transcription start site (TSS). e Cartoon depicting Id2 promoter constructs used for luciferase reporter assay. f Id2 promoter-luciferase construct (−582/+36) were co-transfected with the combination of STAT3, IRF4 and BATF expressing vectors in HEK-293 T cells. Lysates were prepared 30 h after transfection, and luciferase activities were measured with the reporter activities normalized to renilla luciferase activity. *P < 0.05, **P < 0.005, ***P < 0.001 (Student’s t-test). All data are representative three independent experiments with similar results (error bars, s.d.)

Fig. 6

Id2 expression inhibits enrichment and transcriptional activity of E2A on Foxp3 promoter. a RT-qPCR analysis of Foxp3, E2A and Id2 mRNA in in vitro generated T_H0, T_H17, ex-Foxp3 T_H17, and iT_reg cells. b ChIP-qPCR analysis for E2A occupancy at three putative E-box sites on Foxp3 promoter regions (−1593 to −1584, −1295 to −1286, and −837 to −829) and negative control (−411 to −244; No E-box site) in iT_reg and ex-Foxp3 T_H17 cells. Enrichments are calculated relative to the input chromatin for corresponding sites. c Jurkat cells were transfected with either Foxp3 promoter reporter construct alone or along with expression plasmids encoding E2A and Id2 as indicated, followed by luciferase assay. d Effect of mutagenesis on Foxp3 promoter reporter activity. Jurkat cells were transiently co-transfected with the indicated plasmids for luciferase assay as described. N = Normal E box, X = Mutated E box. NS,  not significant, *P < 0.05, **P < 0.005, ***P < 0.001 (Student’s t-test). All data are representative of two or three independent experiments (error bars, s.d.)

Fig. 7

Id2 dependent T_reg plasticity enhances antitumor immunity. a Schematic for doxycycline (Dox)-inducible Id2 overexpression in T_reg cells in TetR-Id2^EmGFPFoxp3^YFP−Cre mice. b Id2 expression was assessed by intracellular staining in SP and pLN from both PBS or Dox treated groups. c Experimental scheme of mouse melanoma model. Females, 8–10 week-old, TetR-Id2^EmGFPFoxp3^YFP−Cre mice were injected subcutaneously with B16.F10 cells. Mice were treated intraperitoneally either with PBS or Dox from day 0–9, every 72 h. d Tumor progression, expressed as mean tumor volume (mm³) in both treatment groups (PBS; n = 8, Dox; n = 8). e Representative tumor size in PBS and Dox treated groups on d 23. f Difference in tumor weight, as measured at end point of analysis (PBS; n = 8, Dox; n = 8). g Flow cytometry analysis of CD4⁺Foxp3⁺ T cells in tumor-draining lymph nodes (dLN), non-draining lymph nodes (non-dLN) as well as within tumor infiltrated lymphocytes. h Intracellular staining of cytokines in CD8⁺ T cells isolated from dLN, non-dLN and tumor infiltrated lymphocytes, stimulated with phorbol myristate acetate (PMA) and ionomycin for 6 h. NS, not significant, *P < 0.05, **P < 0.005, ***P < 0.001 (Student’s t-test). All data are representative of three independent experiments (error bars, s.d.)