Differential Roles of IL-2 Signaling in Developing versus Mature Tregs

Abstract

Although Foxp3⁺ regulatory T cells (Tregs) require interleukin-2 (IL-2) for their development, it has been unclear whether continuing IL-2 signals are needed to maintain lineage stability, survival, and suppressor function in mature Tregs. We generated mice in which CD25, the main ligand-binding subunit of the IL-2 receptor, can be inducibly deleted from Tregs after thymic development. In contrast to Treg development, we find that IL-2 is dispensable for maintaining lineage stability in mature Tregs. Although continuous IL-2 signaling is needed for long-term Treg survival, CD25-deleted Tregs may persist for several weeks in vivo using IL-7. We also observe defects in glycolytic metabolism and suppressor function following CD25 deletion. Thus, unlike developing Tregs in which the primary role of IL-2 is to initiate Foxp3 expression, mature Tregs require continuous IL-2 signaling to maintain survival and suppressor function, but not to maintain lineage stability.

Keywords: CD25; IL-15; IL-2; IL-7; regulatory T cell.

Figure 1.. Loss of Tregs following CD25 Deletion

(A) Sample CD25 gating for CD4⁺ Foxp3⁺ RFP⁺ Tregs showing cutoffs for CD25^hi cells. WT-Tregs may be stratified into WT CD25^hi and WT CD25^lo subsets, while CD25-iΔTregs are defined as CD25⁻ cells. (B) Prevalence of RFP⁺ Foxp3⁺ cells in blood over time, normalized to initial levels measured 4 days after the first tamoxifen injection. (C and D) Proportion of CD25^hi cells among RFP⁺ Foxp3⁺ cells (C) and proportion of Foxp3⁺ cells among CD4⁺ T cells (D) over time. (E) Expression of CD31, a marker of recent thymic emigrants, among RFP⁻ CD25^fl/fl and RFP⁻ CD25^+/+ Tregs 2 weeks after tamoxifen injection. (F and G) BrdU uptake (F) and Annexin-V binding (G) among WT CD25^hi, WT CD25^lo and CD25-iΔTregs 2 weeks following tamoxifen injection. Representative gating is shown to the left of each graph. MFI, mean fluorescent intensity. Values shown are mean ± SD. Data were analyzed using a two-tailed Student’s t test, n = 6 mice (E), or a one-way ANOVA with Tukey’s post-test correction, n = 8–9 mice (F and G). **p < 0.01, ****p < 0.0001. See also Figure S4.

Figure 2.. CD25-iΔTregs Have Reduced Foxp3 Expression but Maintain Lineage Stability

(A) Foxp3 levels among WT CD25^hi, WT CD25^lo, and CD25-iΔTregs following tamoxifen injection. (B) Left: representative gating of RFP⁺ cells to measure Foxp3 expression. Right: Foxp3 histograms, with each line representing a histogram from an individual mouse 10 weeks after tamoxifen administration. (C) Foxp3 expression among WT CD25^hi, WT CD25^lo, and CD25-iΔTregs at 2 weeks (left) and 10 weeks (right) after tamoxifen administration. MFI, mean fluorescent intensity. Values shown are mean ± SD. Data were analyzed using a one-way ANOVA with Tukey’s post-test correction. n = 8–9 mice, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3.. Surface Staining of CD25-iΔTregs Resembles that of WT CD25^lo Tregs

Splenic RFP⁺ Foxp3⁺ Tregs were analyzed by flow cytometry 2 weeks after tamoxifen injection for expression of major Treg surface proteins, including PD-1 (A) and ICOS (B) and CD127, the IL-7 receptor alpha subunit (C). For each marker, representative histograms are shown at top. MFI, mean fluorescent intensity. Values shown are mean ± SD. Data were analyzed using a one-way ANOVA with Tukey’s post-test correction, n = 6–8 mice. *p < 0.05, ***p < 0.001, ****p < 0.0001. See also Figure S5.

Figure 4.. Survival of CD25^hi Tregs following CD25 Deletion

(A) Tamoxifen-untreated CD25^hi or CD25^lo Tregs were sorted from WT-Treg or CD25-iΔTreg donors and transferred into separate CD45.1 ⁺ recipients. Mice were allowed 1 week to rest, injected with tamoxifen, and sacrificed 2 weeks later. (B) Sample plot showing recovery of donor CD45.2⁺ cells. (C) Sample CD25 histograms of recovered CD45.2⁺ Tregs, showing cutoffs for CD25^hi cells. (D) Quantification of CD25^hi cells among recovered Tregs. (E) Percentage of RFP⁺ cells among CD45.2⁺ cells. Representative histogram of RFP expression shown left. (F) Recovery of donor cells following tamoxifen treatment. Values shown are mean ± SD. Data were analyzed using a two-way ANOVA with Tukey’s post-test correction, n = 3–4 mice. **p < 0.01, ****p < 0.0001.

Figure 5.. Reduced Suppressor Function and Glycolysis among CD25-iΔTregs

(A) Sample gating showing dual expression of CD69 and CD154 among 1 × 10⁵ target Foxp3⁻ CD4⁺ T cells in the absence (left) or presence (right) of an equal number of WT-Tregs. (B) Rapid in vitro suppression assay of target cells by titrated numbers of CD25-iΔTregs and WT-Tregs. (C) Clinical severity of EAE in WT-Treg and CD25-iΔTreg mice. Mice were immunized with myelin oligodendrocyte peptide and monitored for 1 month. Tamoxifen was administered at the same time as EAE induction and again 2 weeks later. (D) Representative glycolysis stress test of WT CD25^hi, WT CD25^lo, and CD25-iΔTregs following overnight stimulation in the presence of IL-2. A Seahorse flux analyzer was used to measure extracellular acidification rate (ECAR) of Tregs at baseline and then after treatment with glucose, oligomycin, and 2-deoxyglucose (2-DG) (treatment time, vertical dotted lines). (E and F) Treg glycolytic rate (E) and glycolytic capacity (F) following overnight stimulation in the presence of no cytokine, IL-2, or IL-7. Values shown are mean ± SD. Data were analyzed using a two-tailed Student’s t test, n = 4–5 mice (B) or n = 8–9 mice (C), or a two-way ANOVA with Tukey’s post-test correction, n = 5–6 mice (E and F). *p < 0.05, ***p < 0.001, ****p < 0.0001.

Figure 6.. Concomitant Deprivation of IL-7, but Not IL-15, Exacerbates Treg Loss following CD25 Deletion

(A) Schematic for evaluating efficacy of IL-2, IL-7, and IL-15 at maintaining downstream signals. WT CD25^hi Tregs were stimulated overnight, removed from stimulation, and then incubated with one of three cytokines for another 24 hr. (B and C) Maintenance of downstream phosphorylation by IL-2, IL-7, and IL-15. Bars in representative histograms (B) indicate cutoffs for cells retaining phosphorylation. (D) CD25-iΔTregs or WT CD25^hi Tregs were adoptively transferred into IL-15 KO or IL-7 KO recipients, with an equal number of CD45.1⁺ CD25^hi cells. Recipient mice were sacrificed after 2 weeks. (E and F) Sample gates are shown for recovery of CD45.1⁺ cells (E) and RFP⁺ Tregs (F). (G and H) Recovery of WT-Tregs (G) and CD25-iΔTregs (H) following 2 weeks of IL-15 or IL-7 deprivation in vivo. Values shown are mean ± SD. Data were analyzed using a one-way ANOVA with Dunnett’s post-test correction, n = 3 samples (C), or a one-way ANOVA with Tukey’s post-test correction, n = 4–5 mice (G and H). *p < 0.05, **p < 0.01. See also Figure S6.