Murine CD4⁺CD25⁺ Regulatory T Cells Fail to Undergo Chromatin Remodeling Across the Proximal Promoter Region of the IL-2 Gene

Mice

BALB/c mice were obtained from The Jackson Laboratory (Bar Harbor, ME) and housed under National Institutes of Health-approved guidelines in the Department of Comparative Medicine, Stanford University (Stanford, CA).

Cell preparations and flow cytometry

Isolation by MACS (Miltenyi, Auburn, CA) and FACS, and culture of primary CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells were performed as previously described (10).

IL-2 ELISA and real-time PCR

Equal number of sorted cells was incubated either in the presence of APC, 0.5 μg/ml anti-CD3 (145-2C11) and 1.0 μg/ml anti-CD28 (37.51; BD Pharmingen, San Diego CA), or with 50 ng/ml PMA and 1 μM ionomycin (Ion; Sigma-Aldrich, St. Louis, MO) in 96-well plates. IL-2 level in cultured supernatant was determined by ELISA and IL-2 mRNA levels were determined on cells harvested after 4 h of stimulation using real-time quantitative PCR as previously described (11).

Immunoblot analysis

Following treatment, cells were lysed in buffer containing 20 mM HEPES, pH 7.4, 1% Triton X-100, 100 mM NaCl, 50 mM NaF, 10 mM β-glycerophosphate, 1 mM sodium-vanadate, and 1 mM PMSF. Lysates were centrifuged and protein concentration was determined by Bradford assay (Bio-Rad, Hercules, CA). Proteins were detected with c-Jun (Cell Signaling Technology, Beverly, MA), MKK7 (Zymed, South San Francisco, CA), JNK2 (Santa Cruz Biotechnology, Santa Cruz, CA), and c-Fos (Upstate Biotechnology, Lake Placid, NY). Phospho-specific p38, ERK1/2, and Jun Abs were purchased from Cell Signaling Technology. All blots were developed with HRP-conjugated secondary Abs and ECL.

Lentiviral transduction

MKK7-JNK was cloned into pHR′tripCMV-Ires-tNGFR (12). Lentivirus production in HEK 293T cells and transduction protocols were adapted from a previous report (12). For infection, purified T cells were resuspended at 10 × 10⁶/ml in OptiMEM containing 10 μg/ml protamine sulfate and virus (multiplicity of infection range, 5–10). Cells were incubated at 37°C for 2 h and subsequently were resuspended in complete RPMI containing either IL-7 (20 ng/ml for CD25⁻) or IL-2 (50 U/m) for CD25⁺) to 1 × 10⁶/ml. Forty-eight hours postinfection, nerve growth factor receptor (NGFR)-positive cells were stained with anti-human biotinylated NGFR and streptavidin-APC (BD Pharmingen), sorted and stimulated with PMA/Ion for 12 h.

EMSA

EMSAs was performed according to manufacturer protocol (Geneka, Montreal, Quebec, Canada). Nuclear extracts, prepared from a buffer containing 0.1% Nonidet P-40, 20 mM HEPES, pH 7.4, 10 mM KCl, 1 mM MgCl₂, 400 mM NaCl, and 10% glycerol, were incubated with [γ-³²P]ATP end-labeled consensus wild-type (5′-CGCTTGATGAGTCAGCCGGAA-3′) and mutant AP-1 probes. For supershift assays, indicated Abs were incubated with extracts before the addition of probes.

Chromatin accessibility assay and PCR amplification

Following 4 h of PMA/Ion stimulation, nuclei were isolated as previously described (3). Nuclei were subsequently digested with micrococcal nuclease (Mnase) and subjected to real-time PCR with the following conditions: 95°C for 10 min for one cycle; 95°C for 30 s, 55°C for 1 min, and 72°C for 1 min for 40 cycles using a MX4000 thermocycler (Stratagene, La Jolla, CA). Primer sequences were as follow: SetB: forward, CACAGGTAGACTCTTTGAAAATATGTGTAA, reverse, CATGGGAGGCAATTTATACTGTTAATG; SetD: forward, CTTTTGTGTCTCCACCCCAAA, reverse, CACACTTAGGTGGCAGTTTTAATTCAT; SetF: forward, CATGCAGAGTTTTTTGTTGTTTTCTAG, reverse, GCCTAAAGTCTCTCACAAAGAACAGA; and β-actin: forward, GACGGCCAAGTCATCACTATTG, reverse, AGGAAGGCTGGAAAAGAGCC. Standard curves were generated using genomic DNA, and data were normalized for genomic β-actin. Change in accessibility for individual primer sets was calculated as 1 − [(stimulated)/(unstimulated)].

Highly purified CD4⁺CD25⁺ T_reg cells fail to produce IL-2 upon activation

We previously described a system using latex beads coated with anti-CD3 and anti-CD28 Ab as surrogate APC to study the interaction of CD4⁺CD25⁺ T_reg cells with conventional CD4⁺CD25⁻ T cells in vitro (11). While we could demonstrate loss of suppression in the coculture of CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells with increasing costimulation, the Ab-coated beads did not induce proliferation or IL-2 production in the CD4⁺CD25⁺ T_reg cells. Similarly, highly purified CD4⁺CD25⁺ T_reg cells (Fig. 1⇓A) did not secrete IL-2 when stimulated in vitro with irradiated T cell-depleted splenocytes and soluble anti-CD3/anti-CD28 or with a combination of PMA and Ion (Fig. 1⇓B). We hypothesized that the small amount of IL-2 produced was likely due to nonregulatory CD25⁺ T cells isolated within the T_reg population. Currently, there are no surface markers available for reliable distinction between CD4⁺CD25⁺ T_reg and recently activated conventional CD4⁺CD25⁺ T cells. To estimate the amount of “contaminating” IL-2-producing nonregulatory CD25⁺ T cells in our preparation, we stimulated admixtures of highly purified CD4⁺CD25⁺ T cells and increasing numbers of CD4⁺CD25⁻ T cells with PMA/Ion for 4 h and measured the induction of IL-2 mRNA with quantitative PCR. Data presented in Fig. 1⇓C show a linear relationship between the percentage of admixed CD4⁺CD25⁻ T cells and IL-2 mRNA levels. Based on the intersection of the trend-line with the x coordinate, we estimate that ∼5% of the CD4⁺CD25⁺ T cells are nonregulatory, recently activated T cells, responsible for the minimal IL-2 production seen in the purified T_reg population (Fig. 1⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Lack of IL-2 production in activated CD4⁺CD25⁺ T_reg cells. A, Peripheral CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells were sorted using a combination of MACS and FACS. The purity of the cell preparations was routinely >98% CD25⁺ as indicated by postsort analysis. B, IL-2 production in activated CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells. Cells were stimulated for 24 h with APC and soluble anti-CD3/CD28 (left), or with a combination of 50 ng/ml PMA plus 1 μM Ion (right). Levels of IL-2 in cultured supernatant were measured by ELISA. C, Direct correlation of IL-2 gene expression and percentage of input CD4⁺CD25⁻ T cells. Increasing numbers of CD4⁺CD25⁻ T cells were added to purified CD4⁺CD25⁺ T_reg cells. IL-2 mRNA levels were determined following 4 h PMA/Ion stimulation by real-time quantitative PCR. IL-2 levels of individual samples were normalized to β-actin.

Activated CD4⁺CD25⁺ Treg cells exhibit normal p38 MAPK and ERK1/2 activation and inducible AP-1 binding activity

The MAPK family of proteins mediates a series of distinct signaling cascades that ultimately regulate a multitude of cellular functions including T lymphocyte proliferation and differentiation (13). The three major mammalian MAPK groups, the ERK, p38 MAPK, and JNK kinases, have all been implicated in mechanism(s) of T cell clonal anergy. T cell clones rendered anergic by TCR stimulation in the absence of costimulatory signals have been reported to display defects in ERK, p38 MAPK, and JNK activity (6, 7, 8, 9). The MAPK family of kinases is activated by dual phosphorylation on the threonine (T) and tyrosine (Y) residues at the T-X-Y motif in the activation loop. A time-course treatment with PMA/Ion was performed on freshly isolated CD4⁺CD25⁺ T_reg and CD4⁺CD25⁻ T cells, and lysates were subjected to immunoblot analysis for MAPK activation using dual T/Y phosphorylation-specific MAPK Abs. In both populations of T cells, stimulation with PMA/Ion led to rapid and robust phosphorylation of p38 MAPK and ERK1/2 (Fig. 2⇓A).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

MAPK activation and inducible AP-1 formation in CD4⁺CD25⁺ T_reg cells. A, Similar levels of p38 MAPK and ERK1/2 kinase phosphorylation in activated CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells. Freshly isolated cells were treated with PMA/Ion for the times indicated. Whole cell extracts were subjected to immunoblot analysis with phosphorylation-specific p38 MAPK (top) and p44/42 ERK1/2 (bottom) Abs. B, Similar levels of Fos and Jun expression in activated CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells. Whole cell lysates of T cells were prepared after stimulation for the indicated times and subjected to immunoblot analysis with c-Fos (top) and c-Jun (bottom) Abs. C, Inducible AP-1 DNA binding activity in activated CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells. Freshly isolated cells were treated with PMA/Ion for 2 h. EMSA was performed with 2 μg of nuclear extract incubated with ³²P-labeled consensus AP-1 binding site oligo, in the absence or presence of 100× excess unlabeled wild-type (WT) or mutant (MT) AP-1 oligo. D, A supershift mobility assay reveals similar Jun subunit composition in inducible AP-1 DNA binding complexes. Cells were isolated and treated as in C. In indicated lanes, Abs to JunB and JunD were incubated with nuclear extracts before gel loading.

One downstream effect of ERK pathway activation includes formation and activation of AP-1, a transcriptional dimer complex comprised of Jun and Fos family members. In lymphocytes, AP-1 has been shown to be a key factor in regulating IL-2 transcription. Furthermore, this transcriptional complex has been suggested to be a unique target in T cell clonal anergy (7). In the in vivo staphylococcal enterotoxin A superantigen-induced T cell anergy model, the anergy phenotype has been attributed to defective AP-1 formation, due in part to impaired expression of the Fos and Jun family members (6). To investigate whether differences in Fos and Jun protein levels contributed to defective IL-2 expression in the CD4⁺CD25⁺ T_reg cells, whole cell lysates were prepared and subjected to Western blot analysis. Increased levels of c-Fos protein were detected in both, CD4⁺CD25⁻ and CD4⁺CD25⁺ T cell populations within 30 min of PMA/Ion stimulation, increasing steadily for up to 2 h of treatment (Fig. 2⇑B, top). Similarly, we noted comparable increases in c-Jun protein levels in both cell populations following 2 h of PMA/Ion stimulation (Fig. 2⇑B, bottom). While the induced levels of Fos and Jun protein did not appear to differ between CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells, defects in posttranslational modifications of these subunits could alter AP-1 complex formation. To examine possible perturbations in AP-1 formation, nuclear extracts were prepared from untreated and stimulated cells and subjected to EMSAs. Following 2 h of PMA/Ion treatment, there was no quantitative difference in the level of inducible AP-1 DNA binding activity between CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells (Fig. 2⇑C). Specificity of AP-1 binding activity in both cell populations was demonstrated by competition with an excess of unlabeled wild-type (WT) and mutant (MT) AP-1 probes. Defective formation of NF-AT and NF-κB complexes has also been reported in T cell anergy models (6, 14). However, gel shift assays did not reveal detectable differences in inducible formation of NF-AT and NF-κB complexes between CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells (data not shown).

AP-1 complexes composed of different Fos and Jun family members can contribute to the regulation of cellular proliferation and differentiation. For example, the process of adipocyte differentiation has been shown to be regulated by AP-1 DNA binding activity mediated by c-Jun, JunB, and JunD, and the expression levels of these Jun proteins (15). To study the composition of the inducible AP-1 complex in the CD4⁺ T cells, we conducted supershift EMSA with Abs to specific Jun family members. Incubation of CD4⁺CD25⁻ and CD4⁺CD25⁺ nuclear extracts with specific JunD Ab did not result in a shifted complex, while supershift assays in the presence of specific JunB Ab revealed that the AP-1 complex of both cell populations contained JunB (Fig. 2⇑D). In summary, while perturbations in the formation of activation-dependent transcriptional complexes have been suggested as a potential mechanism for T cell anergy, we did not detect alterations in the formation of the AP-1 DNA binding complex in the activated CD4⁺CD25⁺ T_reg cells.

Activated CD4⁺CD25⁺ T_reg cells display defective JNK activation

While we observed no defects in p38 MAPK and ERK1/2 activation, PMA/Ion treatment resulted in dramatically reduced phosphorylation of JNK1/2 in the CD4⁺CD25⁺ T_reg cells compared with conventional CD4⁺ T cells, though both cell populations displayed comparable levels of JNK protein (Fig. 3⇓A, middle and bottom). In addition, we observed a reduction in MKK4 phosphorylation, one of two upstream MAPK kinases known to activate JNK (16) (Fig. 3⇓A, top).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Impaired JNK signaling pathway does not account for the failure of IL-2 production. A, Impaired activation of JNK signaling pathway in CD4⁺CD25⁺ T_reg cells. Freshly isolated cells were treated with PMA/Ion for times indicated, and whole cell extracts were subjected to immunoblot analysis with phospho-specific MKK4 (top) and JNK1/2 (middle) Abs. Blot was reprobed for JNK2 (bottom). B, Lentiviral transduction of CD4⁺ T cells with constitutively active JNK. Freshly isolated CD4⁺ T cells were transduced with either empty lentiviral vector (dashed line) or with constitutively active MKK7-JNK construct (solid/shaded line). Forty-eight hours posttransduction, cells were analyzed and separated based on biscistronic NGFR surface expression. Transduction efficiency of >50% was observed for both CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells. C, Reconstitution with constitutively active JNK does not abrogate block in IL-2 production in CD4⁺CD25⁺ T_reg cells. Freshly isolated CD4⁺CD25⁻ T cells, and CD4⁺CD25⁺ T cells transduced with either empty vector (VEC) or MKK7-JNK were stimulated for 12 h with PMA/Ion. Levels of IL-2 in cultured supernatant were measured by ELISA. D, MKK7-JNK possesses constitutive JNK activity. HEK 293T cells were transduced with either vector or MKK7-JNK construct. 48 h posttransduction, whole cell lysates were subjected to immunoblot analysis with MKK7 (top) or phospho-specific Jun (middle) Abs. Equal protein loading was verified by probing with ERK1/2 Ab.

The role of the JNK pathway in regulating IL-2 expression has been controversial. A study using a competitive inhibitor of JNK activity implicated the JNK pathway in regulating IL-2 gene transcription (17). Furthermore, anergic Th1 T cell clones have been demonstrated to lack JNK activity upon Ag receptor engagement or stimulation with PMA/Ion (8). However, investigations with JNK1- and JNK2-deficient T lymphocytes revealed a dispensable role for JNK protein kinases in IL-2 production (18). To assess the possible role of JNK in mediating IL-2 production in CD4⁺CD25⁺ T_regs, we chose to introduce a fusion construct of MKK7-JNK into the regulatory T cells. This MKK7-JNK fusion protein contains constitutive JNK activity as shown by phosphorylation of the endogenous substrate, c-Jun, in virally transduced HEK 293T cells (Fig. 3⇑D) and as earlier reported (19). Due to the nonproliferative nature of freshly isolated CD4⁺CD25⁺ T_reg cells, we used a previously described bicistronic lentiviral system (12). Lentiviral transduction of freshly isolated naive CD4⁺CD25- and CD4⁺CD25⁺ T_reg cells resulted in >50% transduction efficiency, based on bicistronic expression of NFGR (Fig. 3⇑B). Forty-eight hours posttransduction, cells transduced with either empty (VEC) or MKK7-JNK (MKK/JNK) vector were sorted and exposed to PMA/Ion for 12 h. Reconstitution of CD4⁺CD25⁺ T_reg cells with MKK7-JNK resulted in increased endogenous c-Jun phosphorylation, as determined by protein microarray analysis (20), but did not result in IL-2 production following PMA/Ion treatment (Fig. 3⇑C). Thus the defect in JNK activation does not provide an explanation for the inability of CD4⁺CD25⁺ T_reg cells to produce IL-2.

Minimal IL-2 promoter of CD4⁺CD25 T_reg cells remains in a closed chromatin configuration upon activation

In addition to integration of signaling pathways and formation of transcriptional complex machinery, chromatin configuration of cell type-specific, stimulation-dependent genes forms an additional level of transcriptional control of cytokine gene expression in mammalian cells. The nucleosomal DNA structure of cytokine genes, not transcribed under basal conditions, is referred to as a “closed” chromatin configuration. Upon extracellular stimulation (i.e., Ag challenge or differentiation cues), areas of high transcriptional activity undergo reconfiguration of the nucleosomal DNA to achieve an “open” configuration, allowing chromatin accessibility to transcriptional complexes. Previous studies have reported changes in chromatin accessibility of the IL-2 locus corresponding to increases in IL-2 gene expression. Nuclease digestion sensitivity assays have revealed increased accessibility within the proximal minimal IL-2 promoter upon activation in primary T cells and T cell clones, indicative of a change in chromatin configuration (2, 3, 21).

To investigate whether the proximal promoter region of the IL-2 gene in murine CD4⁺CD25⁺ T_reg cells becomes accessible to nuclease digestion upon activation, we used a recently described chromatin accessibility by real-time PCR (CHART-PCR) assay (3). In brief, nuclei isolated from unstimulated and stimulated cells are treated with MNase, a DNase that preferentially digests nucleosome-free regions of DNA. The extent of nucleosome remodeling as detected by MNase accessibility can be quantified by real-time quantitative PCR across the region of interest. Accessibility is inversely proportional to the amount of PCR product (threshold cycle value) obtained using primers encompassing the perturbed region of DNA. The CHART-PCR assay was particularly suited for our purposes as it can be performed with relatively few numbers of cells.

The positions of the PCR fragments relative to the IL-2 transcriptional start site we investigated are shown in Fig. 4⇓A. The nonoverlapping primer sets B (−115 to −14 bp), C (−200 to −110 bp), and D (−330 to −230 bp) cover a region that was reported to show increased MNase accessibility upon PMA/Ion stimulation of conventional T cells, while no change was detected in the distal region, F (3). To validate the sensitivity of this in vivo chromatin remodeling assay, nuclei isolated from unstimulated and PMA/Ion treated CD4⁺CD25⁻ and CD4⁺CD25⁺ cells were subjected to MNase treatment, followed by real-time PCR. Using primers spanning region D, a clear shift is observed in PMA/Ion-stimulated conventional CD4⁺ T cells, compared with untreated cells, indicative of an “open” configuration in this region. Interestingly, there is no observable difference in MNase accessibility between unstimulated and stimulated CD4⁺CD25⁺ T_reg cells (Fig. 4⇓B, left). To demonstrate regional selectivity, no change is observed in the distal region F in either subpopulation of cells (Fig. 4⇓B, right).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

The IL-2 promoter region retains a closed chromatin configuration in CD4⁺CD25⁺ T_reg cells. A, Schematic diagram of murine IL-2 promoter/enhancer region. Primer sets used for accessibility assay are indicated, with transcriptional start site represented by (+1). B, Inaccessibility of selected IL-2 promoter regions in CD4⁺CD25⁺ T_reg cells. Nuclei of untreated CD4⁺CD25⁻ (□) and CD4⁺CD25⁺ (▵), and PMA/Ion-stimulated CD4⁺CD25⁻ (▪) and CD4⁺CD25⁺ (▴) T cells were treated with MNase for 5′ and genomic DNA was subjected to SYBR Green real-time PCR. CD4⁺CD25⁻ and CD4⁺CD25⁺ T cell samples were subjected to real-time quantitative PCR with primers spanning region D (left) and with primers spanning region F (right). C, Proximal IL-2 promoter of CD4⁺CD25⁺ T_reg cells does not undergo change in chromatin accessibility upon activation. Nuclei of stimulated CD4⁺CD25⁻ and CD4⁺CD25⁺ T cells were treated and subjected to real-time quantitative PCR as described in B using primers spanning indicated regions. Standard curves for individual primer sets were generated using serial dilutions of genomic DNA. Values were normalized for genomic β-actin. The data are plotted as the percentage of change in accessibility following 4 h of PMA/Ion stimulation. Mean data of four separate experiments are shown.

RT-PCR data of tested regions, normalized to an internal actin control, are shown in Fig. 4⇑C. Conventional CD4⁺ T cells treated with PMA/Ion display >60% increase in chromatin accessibility in the regions covered by primer sets B, C, and D. In contrast, no change in accessibility of these regions is observed in the CD4⁺CD25⁺ T_regs following PMA/Ion treatment. These data strongly suggest that the proximal IL-2 promoter in CD4⁺CD25⁺ T_reg cells remains in a closed nucleosomal structure upon activation and does not undergo the selective chromatin remodeling observed in conventional CD4⁺ T cells following similar activation.