Mice

CD45.1 congenic (B6.SJL-Ptprc^a Pepc^b/BoyJ), Rag1 deficient (Rag1^−/−, CD45.2, B6), IL2 deficient (Il2^−/−, B6.129P2-Il2^tm1Hor/J, CD45.2) and Foxp3-GFP knock-in (Foxp3 bicistronic reporter mice expressing EGFP: B6.Cg-Foxp3^tm2Tch/J, CD45.2) mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). Mice were all in the C57BL/6 (B6) background and bred in the animal facility at the Hallym University. Since the phenotypes of Il2^−/− mice are characterized by prominent inflammatory responses (Schorle et al. 1991) and these phenotypic changes can affect the physiology of Tregs (Fontenot et al. 2005), we tried to use young wild type (WT) and Il2^−/− mice where inflammation did not start yet or was mild (8−15 days of age unless otherwise specified in the text or figure legends). Foxp3-GFP knock-in allele(s) was introduced into all genetically modified mice by mating and GFP signals were used to identify Tregs. Animal experimentations were conducted in accordance with the guidelines of the Experimental Animal Center at Hallym University (Hallym 2012-61-1, 2013-109).

Cell isolation and flow cytometry

To sort peripheral T cell subpopulations from spleens and lymph nodes (LNs), the CD4⁺ fraction was first purified on the MACS using anti-CD4 beads (Miltenyi Biotec, Bergisch Gladbach, Germany). The enriched CD4⁺ fractions were then further separated into each subpopulation by using FACS Aria-II flow cytometer (BD Biosciences, San Jose, CA, USA). To sort naïve CD4⁺ cells and peripheral Treg cells, CD44 and GFP were used (naïve cells: CD4⁺CD8⁻GFP⁻CD44^low; Treg cells: CD4⁺CD8⁻GFP⁺). The post-sort purity for each cell type was usually >95%.

Cell culture

FACS-sorted cells were cultured in complete RPMI-1640 medium (WelGENE, Daegu, Korea), supplemented with 10% fetal bovine serum (WelGENE), penicillin, streptomycin (Sigma-Aldrich, St. Louis, MO, USA), L-glutamine (2 mM; Life Technologies, Carlsbad, CA, USA), sodium pyruvate (2 mM; Sigma-Aldrich), nonessential amino acid (0.1 mM; Sigma-Aldrich) and 2-ME (50 μM; Sigma-Aldrich) (Heo et al. 2016).

Colitis model

Colitis was induced by FACS-purified naïve CD4⁺ T cells (CD4⁺CD44^lowCD45RB^highGFP⁻) either alone or in combination with Treg cells (CD4⁺GFP⁺) having different congenic marker (CD45.1 or CD45.2). Mice were weighed and monitored regularly for the signs of disease. Mice were sacrificed at 10 weeks after T cell transfer and their colons were used for histology.

Histology

Complete colons were fixed in formalin, processed, and stained with hematoxylin and eosin (H & E) or Periodic acid-Schiff (PAS). Blinded sections from the entire colon were examined.

Administration of anti-CD40 mAb

Mice were injected i.p. with 10 μg anti-CD40 (IC10; eBioscience) or rat IgG on days 0, 2, and 4 and sacrificed on day 6. Pooled peripheral LNs were analyzed by using the flow cytometry.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) analyses were done using EZ-ChIP kit (Milipore, Darmstadt, Germany). In brief, chromatin prepared from purified naive CD4⁺ T cells and CD4⁺GFP⁺ Treg cells was pre-cleared for 1 h with protein A agarose beads and then was incubated overnight with 2 μg rabbit anti-STAT5 (Santa Cruz Biotechnology, Dallas, TX, USA) or control rabbit anti-MyD88 (Santa Cruz Biotechnology). Immune complexes were precipitated with protein A agarose beads. Precipitated DNA was analyzed by real time PCR (Kim et al. 2014).

The sequences of primers are as below:

Tbx21 enhancer-Forward: 5’-CTTAGAAGGGGGTGGGTAGC-3’

Tbx21 enhancer-Reverse: 5’-GGACTGGAAAATCAGGCTCA-3’

Statistical analyses

A two-tailed, unpaired, student’s t-test was used to calculate the statistical significance of differences between groups unless specified. P values were represented as follows: ***P < .001; **P < .01; *P < .05, whereas NS, not significant, was used to denote P values >.05. Error bars indicate s.e.m.

The expression of CXCR3 was reduced in Il2^−/− Tregs

While we examined various surface and intracellular molecules in WT and Il2^−/− Tregs with flow cytometry, we found that Il2^−/− Tregs showed naïve phenotypes such as CD25^low, CD103^low, CTLA4^low and CD44^low in contrast to WT Tregs showing activated phenotypes (Figure 1(a)). Among these markers, we were interested in CXCR3, a chemokine receptor related with Th1 immune responses, which was downregulated in Il2^−/− Tregs but not in conventional CD4⁺ T cells (Figure 1(b)). Consistent with a previous report (Koch et al. 2009), the chemokine receptor CXCR3 was found in about 15−20% of the WT Tregs. However, the frequency of CXCR3⁺ cells was significantly lower in Il2^−/− Tregs (Figure 1(c)). The lack of CXCR3⁺ Tregs in Il2^−/− mice suggests that CXCR3 expression is directly controlled by IL2, or there might be intrinsic defects in Th1 signaling molecules in Il2^−/− Tregs. To distinguish these possibilities, we did adoptive transfer experiments. Tregs FACS-sorted from LNs of WT mice were injected into congenically marked Il2^−/− mice and analyzed after 7 days. Flow cytometric analysis revealed that the frequency of CXCR3⁺ cells in the donor WT Tregs was still higher even in the Il2^−/− host mice (Figure 1(d)) suggesting that Il2^−/− Tregs have an intrinsic defect in CXCR3 expression. Because IFN-γ receptor (IFN-γR) and STAT-1 are known to promote the expression of CXCR3 (Koch et al. 2009), we compared the expression levels of CXCR3 in IL2, IFN-γR and double deficient Tregs. Indeed, the frequency of CXCR3⁺ Tregs was reduced the most in double deficient mice (Figure 1(e)). Furthermore, the expression level of IFN-γR was not altered in Il2^−/− Tregs (Figure 1(f)) implying that CXCR3 was downregulated in the Il2^−/− Tregs in an IFN-γR−independent and cell-intrinsic manner.

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 1. The expression of CXCR3 in Tregs is dependent on IL2. (a) Phenotypic analysis of Tregs in WT and Il2^−/− mice. FACS plots are gated on CD4⁺Foxp3⁺ cells. (b) The frequencies of CXCR3⁺ cells in each population are shown. Gray, filled histograms are the results of the isotype control. (c) Results of the statistical analysis of (b), Bars show the mean ± SEM (n = 6−8, pooled from three independent experiments). (d) Congenically labeled WT Tregs were FACS-sorted and injected into Il2^−/− mice. After 7 days, CXCR3 expression of the Tregs of donor or host origin was analyzed separately by flow cytometry. (e) CXCR3 expression in the Tregs of WT, Il2^−/− (IL2KO), IFNγR deficient (Ifngr^−/−, IFNγR-KO) and double deficient (Il2^−/−Ifngr^−/−, DKO) mice. (f) Flow cytometry showing the percentage of IFNγR-expressing cells in the CD4⁺ and Treg populations from WT and Il2^−/− mice. Data are representative of three independent experiments (a, b, d, e, f). Numbers in the indicated area in the FACS plots refer to the percentage of each subset. ***P < .001; *P < .05; NS, not significant.

Display full size

Il2^−/− Tregs failed to upregulate T-bet under Th1 inflammations

Because T-bet has an essential role in CXCR3 expression in Tregs (Koch et al. 2009), we sought to investigate whether IL2 is involved in regulating the expression of T-bet in Tregs. T-bet is upregulated by specific cues associated with Th1 responses, and strong Th1 responses can be induced in vivo simply by injecting an agonistic anti-CD40 mAb (Ferlin et al. 1998), which led us to examine T-bet expression in WT and Il2^−/− mice treated with anti-CD40 mAbs. As expected, the frequency of T-bet−expressing Tregs was increased in both WT and Il2^−/− mice treated with anti-CD40 mAbs. However, the extent of upregulation of T-bet was much smaller in the Il2^−/− Tregs (Figure 2(a)). We also did a FACS analysis on the expression of CXCR3 and found similar patterns (Figure 2(b)).

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 2. T-bet and CXCR3 were not upregulated in Il2^−/− Tregs after anti-CD40 mAb treatment. Flow cytometric analysis of T-bet (a) and CXCR3 (b) in the Tregs of WT or Il2^−/− mice treated with rat IgG (control) or anti-CD40 mAb. The first FACS plots are the isotype control samples. Data are representative of three independent experiments. Numbers in the indicated area in the FACS plots refer to the percentage of each subset.

Display full size

IL2-STAT5 signals control the expressions of T-bet directly in Tregs

Next, we analyzed T-bet expression in Tregs after rIL2 treatment. Purified WT CD4⁺ naïve T cells and Tregs were cultured in the presence of rIL2 (2000 U/ml) overnight, and tbx21 transcripts were checked by quantitative PCR (qPCR). The level of tbx21 was not changed by the rIL2 treatment in naive CD4⁺ T cells (data not shown). In contrast, tbx21 expression in the Tregs increased after treatment with rIL2 (Figure 3(a)). Previously, IL2 was reported to facilitate Th1 cell differentiation by inducing T-bet in activated CD4⁺ T cells (Liao et al. 2011), which led us to examine whether IL2 upregulates T-bet directly via STAT5 in Tregs. Indeed, there was significantly more STAT5 bound to the tbx21 promoter locus in IL2−treated Tregs (Figure 3(b)). Altogether, these results imply that IL2 increases the expressions of T-bet directly via STAT5 enabling a subset of Tregs to be equipped with Th1 associated molecules including CXCR3.

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 3. The expression of T-bet in Tregs is controlled by the IL2−STAT5 signaling pathway. (a) FACS-sorted WT Tregs were treated with rIL2 (2000 U/ml) overnight in vitro, and RNAs extracted from each were subjected to qPCR analysis. The values were normalized to actin. The mean values ± SEM fold changes are shown. (b) ChIP analysis of STAT5 binding to the tbx21 promoter locus in WT naïve CD4⁺ T cells or Tregs cultured overnight with media (−) or rIL2 (+, 2000 U/ml). Each sample was immunoprecipitated with anti-STAT5 antibody, and bound DNA was amplified by qPCR for tbx21. Data are representative of three independent experiments. *P < .05; NS, not significant.

Display full size

We next analyzed Th2 transcription factor GATA3 to examine whether IL2 has broader effects on the expressions of helper T cell lineage transcription factors, as shown in helper T cell differentiation (Liao et al. 2008; Liao et al. 2011), and found that GATA3 was also downregulated in Il2^−/− Tregs (Figure 4). These results led us to hypothesize that IL2 might be required for genetic heterogeneity in Tregs (Campbell & Koch 2011) and help Tregs hold corresponding immune responses in check.

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 4. IL2 is essential for GATA3 expression in Tregs. GATA3 expression in WT and Il2^−/− (KO) CD4⁺Foxp3⁻ conventional T cells or CD4⁺Foxp3⁺ Tregs. Data are representative of three independent experiments. Numbers in the indicated area in the FACS plots refer to the percentage of each subset.

Display full size

Impaired suppressive activity of Il2^−/− Tregs

To examine the suppressive activity of Il2^−/− Tregs in vitro, we did the in vitro suppression test. Cell division dye (eFluoro670)-labeled CD4⁺ responder cells were cocultured with antigen presenting cells, Tregs, and anti-CD3 mAbs, and the extent of proliferation of the responders was estimated after 3 days with flow cytometry. Because the dyes remain in non-dividing cells and are diluted in dividing cells, the frequency of dye⁺ responders indicates the suppressive activity of the Tregs. FACS-sorted Tregs (WT or Il2^−/−) and WT naïve CD4⁺ T cells (responders) were mixed and stimulated by anti-CD3 mAbs in the presence of antigen presenting cells, and the proliferation of the responder cells was analyzed. Although the responders could produce IL2 and provide it to the Il2^−/− Tregs, the suppressive function of the Il2^−/− Tregs was impaired (Figure 5). Collectively, these results suggest that the immune regulatory functions of the Il2^−/− Tregs might not be optimal especially under Th1 immune responses.

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 5. Analysis of suppressive functions in Il2^−/− (KO) Tregs Responders were labeled with eFluoro670 and cultured in the presence or absence of anti-CD3 mAb and various doses of Tregs. Fluorescent intensity of eFluoro670 in responders was detected by flow cytometry after 3 days. Results of statistical analysis (right) are from a single experiment (duplicates, mean ± SEM) representative of three independent experiments.

Display full size

Il2^−/− Tregs failed to prevent colitis

Next, we analyzed the suppressive functions of Il2^−/− Tregs in the colitis model (Haribhai et al. 2009). WT naïve CD4⁺ cells (5 × 10⁵, CD45.1⁺CD45.2⁺) were transferred to Rag1 deficient mice either alone or in combination with 1.5 × 10⁵ Tregs FACS-sorted from WT (CD45.1⁻CD45.2⁺) or Il2^−/− (CD45.1⁻CD45.2⁺) mice (Figure 6(a)). The expression of Foxp3 was examined by intracellular staining, and >95% of the sorted Tregs expressed Foxp3 (Figure 6(a), lower). Recipient mice were sacrificed and analyzed by histological examination (for colonic inflammation) and flow cytometry (for phenotypic analysis) in 8−10 weeks. As described previously (Mottet et al. 2003), transfer of naïve CD4⁺ T cells resulted in clinical signs such as weight loss and histological changes. These abnormalities were ameliorated by co-transfer of WT Tregs. In contrast, mice that received Il2^−/− Tregs developed colitis (Figure 6(b)) and intestinal inflammation (Figure 6(c)). Like the previous report that proinflammatory cytokine IFN-γ was associated with the pathogenesis of colitis (Xavier & Podolsky 2007), Il2^−/− Tregs could not suppress IFN-γ production in pathogenic T cells (Figure 6(d)). To determine the mechanisms behind these results, we separated T cells according to the expression of congenic markers (CD45.1 and CD45.2) and analyzed them with flow cytometry. In both WT and Il2^−/− Tregs, the expression levels of CD25, Foxp3 and Ki67 were not different (data not shown). However, the frequency of CD45.1⁻CD45.2⁺ cells of Il2^−/− Treg origin was much lower (Figure 6(e)) suggesting that the survival of the Il2^−/− Tregs was impaired. Foxp3 proteins were also lost in some Il2^−/− Tregs although the difference was not statistically significant (Figure 6(f)).

IL2 is required for functional maturation of regulatory T cells

All authors

Phil Hyun Jeon & Kwon Ik Oh

https://doi.org/10.1080/19768354.2016.1272489

Published online:

30 December 2016

Figure 6. Il2^−/− Tregs did not protect mice from colitis. (a) FACS-sorted WT naïve (CD45.1⁺CD45.2⁺), Tregs (CD45.1⁻CD45.2⁺) and Il2^−/− Tregs (CD45.1⁻CD45.2⁺) were injected alone or together into Rag1 deficient (Rag^−/−) mice. Recipient mice were examined after 8−10 weeks (left). FACS analysis scheme for the recipient mice is on the right side. (b) Rag1 deficient mice received either naïve T cells alone (black) or in combination with WT (white) or Il2^−/− (KO, gray) Tregs, and were weighed weekly. The weight changes are shown as the mean ± SEM (n = 8, pooled from two independent experiments). (c) Representative histological sections of the colons from the mice in (b) stained with Hematoxylin and Eosin (H-E, upper) and Alcian blue and PAS (lower) to visualize goblet cells and mucus deposition. (d) Intracellular staining for IFN-γ in CD4⁺ T cells of naïve T cell origin (CD45.1⁺CD45.2⁺) after PMA/ionomycin stimulation. (e) The percentage of cells of Treg origin (CD45.1⁻CD45.2⁺) in total transferred CD4⁺ T cells (CD4⁺TCRβ⁺). Results of statistical analysis are shown on the right side. Each symbol represents the result obtained from an individual recipient mouse; small horizontal lines indicate the mean values. (f) The percentage of Foxp3⁺ cells in the CD45.1⁻CD45.2⁺ population (Treg origin). Results of statistical analysis are shown on the right side. Each symbol represents an individual recipient mouse; small horizontal lines indicate the mean values. mLN, mesenteric LN. Numbers in the indicated area in the FACS plots refer to the percentage of each subset. ***P < .001; *P < .05; NS, not significant.

Display full size