Skip to main content
PMC home page
Immunology logoLink to Immunology
. 2014 Feb 10;141(3):362–376. doi: 10.1111/imm.12198

Histone methylation mediates plasticity of human FOXP3+ regulatory T cells by modulating signature gene expressions

Haiqi He 1,2,*, Bing Ni 3,*, Yi Tian 3,*, Zhiqiang Tian 3, Yanke Chen 4, Zhengwen Liu 2,5, Xiaomei Yang 6, Yi Lv 1,2,✉, Yong Zhang 7
PMCID: PMC3930375  PMID: 24152290

Abstract

CD4+ FOXP3+ regulatory T (Treg) cells constitute a heterogeneous and plastic T-cell lineage that plays a pivotal role in maintaining immune homeostasis and immune tolerance. However, the fate of human Treg cells after loss of FOXP3 expression and the epigenetic mechanisms contributing to such a phenotype switch remain to be fully elucidated. In the current study, we demonstrate that human CD4+ CD25high CD127low/− Treg cells convert to two subpopulations with distinctive FOXP3+ and FOXP3− phenotypes following in vitro culture with anti-CD3/CD28 and interleukin-2. Digital gene expression analysis showed that upon in vitro expansion, human Treg cells down-regulated Treg cell signature genes, such as FOXP3, CTLA4, ICOS, IKZF2 and LRRC32, but up-regulated a set of T helper lineage-associated genes, especially T helper type 2 (Th2)-associated, such as GATA3, GFI1 and IL13. Subsequent chromatin immunoprecipitation-sequencing of these subpopulations yielded genome-wide maps of their H3K4me3 and H3K27me3 profiles. Surprisingly, reprogramming of Treg cells was associated with differential histone modifications, as evidenced by decreased abundance of permissive H3K4me3 within the down-regulated Treg cell signature genes, such as FOXP3, CTLA4 and LRRC32 loci, and increased abundance of H3K4me3 within the Th2-associated genes, such as IL4 and IL5; however, the H3K27me3 modification profile was not significantly different between the two subpopulations. In conclusion, this study revealed that loss of FOXP3 expression from human Treg cells during in vitro expansion can induce reprogramming to a T helper cell phenotype with a gene expression signature dominated by Th2 lineage-associated genes, and that this cell type conversion may be mediated by histone methylation events.

Keywords: chromatin immunoprecipitation-sequencing, conversion, regulatory T cells, transcriptome

Introduction

Natural CD4+ regulatory T (Treg) cells play an indispensable role in the maintenance of immune tolerance and immune homeostasis.1 Adoptive transfer of Treg cells has been suggested as a potential therapy for the management of autoimmune diseases and the prevention or treatment of alloresponses after organ or stem cell transplantation.1 In animal models, adoptive transfer of Treg cells has been shown to prevent or even cure autoimmunity.2–4 More recently, adoptive transfer of Treg cells in human patients has shown this type of cellular therapy as an effective strategy to prevent graft-versus-host disease after stem cell transplantation.5–7 Despite the considerable enthusiasm generated by these clinical studies, widespread application of Treg cell therapy has been hampered by the unresolved technical challenges to preparing and expanding a functionally stable population in vitro for safe and effective in vivo usage.1

Our current knowledge of the origin of Treg cells suggests that this cell population is composed of two potentially distinct subpopulations: thymus-derived (t)Treg cells that differentiate in the thymus and peripherally derived (p)Treg cells that differentiate in the periphery.8 Additionally, Treg cells can also be generated in vitro from the naive T cells by a variety of means [for example, through activation in the presence of transforming growth factor-β and interleukin-2 (IL-2)], and are designated accordingly as in vitro-induced (i)Treg cells.9 All subpopulations, however, are characterized by the expression of FOXP3, a master transcription factor that exerts its activator and suppressor activities to establish the transcriptional network critically involved in Treg cell development and function.10–12 Moreover, the transcriptional programmes underlying Treg cell function seem to depend on stable and continuous FOXP3 expression, as attenuating Foxp3 expression or deleting the Foxp3 gene in mature murine Treg cells results in loss of their immunosuppressive function.13,14

Genotype–phenotype analyses have also suggested that a greater extent of heterogeneity exists in the human Treg cells, with many phenotypically and functionally distinct subpopulations present among the FOXP3+ cells.15 For example, studies of these cells based on expression status of CD45RA have characterized the robust immunosuppressive activity of CD45RA− Treg cells (designated as a memory-type Treg cell), and defined the CD45RA+ Treg cell subset (naive-type) as an optimal candidate for in vitro expansion.15,16 Yet, it has been noted that upon in vitro expansion, the FOXP3+ Treg cells lose their FOXP3 expression and acquire effector T helper (Th) cell functions.17,18 Studies of this reprogramming process have implicated Th cell polarizing cytokines or repetitive stimulation of the T-cell receptor (TCR)-mediated signalling pathway as contributing aetiologies.17,19–21 Importantly, studies of various in vivo models have also demonstrated the conversion of Treg cells into functional effector Th cells capable of producing the normal panel of pro-inflammatory cytokines, including interferon-γ, IL-2 and IL-17, in particular under the inflammatory or lymphopenic environments;22–24 however, the fate of human Treg cells after loss of FOXP3 expression and the underlying mechanisms of this reprogramming remain undefined.

Previous studies have shown that DNA methylation is crucial for controlling expression of the FOXP3 locus, as evidenced by differential DNA methylation status within the FOXP3 locus of Treg and conventional T (Tconv) cells.25–27 This notion was further supported by the observation of DNA methyltransferase inhibitors inducement of strong Foxp3 expression and increased Treg cell numbers.28 The Treg-specific demethylation region within the FOXP3 gene was defined as a conserved non-coding region that displays complete demethylation in tTreg cells but not in iTreg cells, which only transiently express FOXP3 after activation, and other T cells.29 Interestingly, the Treg-specific demethylation region within the FOXP3 locus in Treg cells was found to be remethylated after loss of FOXP3,17,24 suggesting an essential role for epigenetic modifications in controlling the stability of Treg cells.

Histone modifications are another epigenetic mechanism that affects gene transcription by altering the chromatin structure and DNA accessibility. Histone acetylation is typically associated with open chromatin status and active gene transcription, while histone methylation can be associated with either open or compacted chromatin status. For example, trimethylation of H3K4 and H3K36 and monomethylation of H3K27 and H3K9 are associated with transcriptionally active genes, whereas trimethylation of H3K27 and H3K9 are associated with transcriptionally silenced genes.30–32 It was shown in mice that deacetylase inhibition induced by administration of a histone/protein deacetylase inhibitor leads to an increase in Foxp3 gene expression in CD4+ CD25− and CD4+ CD25+ T cells.33,34 Furthermore, inhibition of histone/protein deacetylase activity has been shown to prevent the conversion of Treg cells into IL-17-producing cells.21 Collectively, these observations suggest that an epigenetic mechanism may contribute to the loss of FOXP3 expression and the reprogramming of Treg cells.

In this study, we found that upon in vitro expansion, human Treg cells diverged into two distinct FOXP3 subpopulations, those that maintained the FOXP3 expression and those that lost their FOXP3 expression. Comparative analysis of transcriptome data from high-throughput digital gene expression (DGE) and histone modification data from chromatin immunoprecipitation-sequencing (ChIP-Seq) provided novel insights into this reprogramming event, indicating that human Treg cells can convert into Th-like cells displaying a gene expression signature dominated by Th2 lineage-associated genes and that histone methylation may contribute to this conversion.

Materials and methods

Isolation and in vitro expansion of human Treg cells

Peripheral blood mononuclear cells were obtained from leukapheresis products of healthy volunteers and isolated by density gradient centrifugation over Ficoll-Paque PLUS medium (GE Healthcare, Pittsburgh, PA); all donors provided informed consent, and the sample collection and study were approved by the ethics committee of Xi'an Jiaotong University. The CD4+ T-cell fraction of the peripheral blood mononuclear cells was enriched using the MidiMACS separator and accompanying reagents from the human CD4+ T-cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). The enriched CD4+ T cells were then stained with CD4-fluorescein isothiocyanate, CD25-phycoerythrin (each in PBS with 2% fetal bovine serum; both from BD Biosciences, San Jose, CA) and CD127-Peridinin chlorophyll protein-Cy5.5 (eBioscience, San Diego, CA) and applied to a FACSAria high-speed cell sorter (BD Biosciences) to sort the CD4+ CD25high CD127low/− and CD4+ CD25− CD127+ subpopulations (Fig. 1a and see Supplementary material, Fig. S1). The purity of sorted cells was re-analysed. The frequency of FOXP3+ cells was analysed by flow cytometry after fixing and permeabilization (solutions from eBioscience) and intracellular staining with FOXP3-allophycocyanin (eBioscience).

Figure 1.

Figure 1

Open in a new tab

Isolation and in vitro expansion of human CD4+ CD25high CD127low/− regulatory T (Treg) cells. (a) MACS-enriched CD4+ T cells from peripheral blood mononuclear cells of healthy human donors were stained with CD4-fluorescein isothiocyanate, CD25-phycoerythrin, and CD127-Peridinin chlorophyll protein-Cy5.5, and sorting gates were applied (left panel) to isolate CD4+ CD25high CD127low/− Treg cells. Re-analysis of sorted CD4+ CD25high CD127low/− T cells is shown in the middle panel. Sorted Treg cells were fixed, permeabilized, intracellularly stained with FOXP3-allophycocyanin, and analysed by flow cytometry (right panel). (b) Treg cells were expanded in vitro for 6 weeks, and cell numbers were determined weekly. Data points represent mean values from four different cultures. (c) and (d) Proportion of FOXP3+ cells in the propagated cells during in vitro expansion. (c) Summarized data from cell cultures initiated from four different leukapheresis products obtained from four different donors. The data are presented as mean ± SD.(d) Representative example of flow cytometry analysis in the propagated cell population.

The isolated cells were expanded in vitro over a period of 5–6 weeks using the Dynabeads® Human Treg Expander superparamagnetic beads (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. Briefly, the Treg cells were cultured in X-vivo 15 medium (Lonza, Walkersville, MD) supplemented with 5% human AB serum (Sigma-Aldrich, St Louis, MO) and then stimulated with anti-CD3/CD28 antibody-coated beads (Invitrogen) in the presence of high-dose recombinant human IL-2 (500 U/ml; PeproTech, Rocky Hill, NJ). During the expansion, the percentage of FOXP3+ cells was analysed every week by flow cytometry as described above. The viable cell analysis was performed weekly by flow cytometry with 7-amino-actinomycin D (BD Biosciences) according to the manufacturer's protocol.

Cell sorting of CD4+ FOXP3+ and CD4+ FOXP3− subpopulations

The expanded Treg cells were sorted into CD4+ FOXP3+ and CD4+ FOXP3− subpopulations on the FACSAria high-speed cell sorter. Briefly, the cells were stained with CD4-fluorescein isothiocyanate and CD25-phycoerythrin, fixed and permeabilized, and stained with FOXP3-allophycocyanin before sorting. All antibodies were dissolved in sterile PBS with 0·1% diethyl pyrocarbonate (Sigma-Aldrich) supplemented with recombinant RNasin® ribonuclease inhibitor (Promega, Madison, WI). The two lineages are hereafter referred to as FOXP3-maintaining and FOXP3-losing Treg cells.

DGE and differential expression analysis

Total RNA was extracted from fixed cells using High pure FFPE RNA Micro Kit (Roche Diagnostic Corp., Indianapolis, IN) and submitted to BGI-Shenzhen (China) for lllumina sequencing. Briefly, mRNA was first isolated from the total RNA by using Oligo(dT) magnetic beads and then applied as template for cDNA synthesis. The resultant cDNA was endonuclease digested (NlaIII), which recognizes and cuts the most 3′ ‘CATG’, and ligated to Illumina adapter 1. Following a sequential digestion of the 5′ end with MmeI and ligation to Illumina adapter 2, the DNA fragments (library) were amplified by PCR, gel purified and sequenced on the HiSeq™ 2000 system.

Raw image data were transformed by base calling into raw sequence data and then transformed into clean tags for alignment to the human reference genome hg19. Only tags with perfect matching or 1 bp mismatch in a single gene were selected for further analysis. The total number of unambiguous clean tags mapping to each gene was normalized to determine the number of transcripts per million clean tags.35,36 All DGE data from the study were submitted to NCBI's Gene Expression Omnibus (GEO; GSE47636, http://www.ncbi.nlm.nih.gov/geo/).

The genes with significant differential expression levels between the two samples were identified according the following criteria37: false discovery rate of ≤ 0·001 and fold-change of ≥ 2. The set of significantly differentially expressed genes was subjected to gene ontology enrichment analysis using the online gene ontology tool (http://www.geneontology.org) and to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.38

ChIP-Seq and genome mapping

The initial ChIP processing was performed as previously described,39,40 with minor modifications. Briefly, a total of 1 × 107 cells were cross-linked with 1% formaldehyde, and chromatin was sonicated to obtain an average fragment length range of 200–500 bp. After pre-clearing with protein A agarose beads (Upstate, Temecula, CA), the sonicated chromatin was precipitated with anti-H3K4me3 (Abcam, Cambridge, UK), anti-H3K27me3 (Abcam), or anti-rabbit IgG (Upstate) by incubating overnight at 4°. The immune complexes were then bound to protein A agarose beads (Upstate), washed, eluted and heated overnight at 65° to reverse the cross-links. The eluted DNA was then treated sequentially with Proteinase K and RNase A, and purified with the QIAquick PCR Purification Kit (Qiagen, Valencia, CA). Before proceeding to the ChIP-Seq procedure, the enrichment efficiency of ChIP was detected by quantitative PCR.

For Illumina ChIP-Seq, the purified DNA fragments were repaired using polynucleotide kinase and Klenow enzyme, ligated to Illumina adapters, amplified by PCR (100–300 bp average lengths) and submitted to BGI-Shenzhen for sequencing on the HiSeq™ 2000 system. All ChIP-Seq data from the study were submitted to NCBI's GEO (GSE47510, http://www.ncbi.nlm.nih.gov/geo/).

Sequenced reads of 49 bp obtained with the Solexa Analysis Pipeline of the Illumina HiSeq™ 2000 system were mapped to the human genome (hg19) by SOAP 2.21, and only alignments with two or fewer mismatching bases were retained.41,42 The output of the SOAP analysis data was converted to browser-extensible data files for viewing the data in the University of California, Santa Cruz genome browser (http://genome.ucsc.edu/).

Identification of H3K4me3 and H3K27me3 peaks, lineage-specific peaks and peak-associated genes

To identify genome areas enriched with a specific histone modification (peaks), the defined analysis model, Model-based Analysis of ChIP-Seq peak-finding algorithm based on Poisson distribution, was used with default parameters.43 To estimate the distribution of H3K4me3 and H3K27me3 modifications, the genome was divided into exons, introns, intergenic regions, up20K [from 20 kb upstream of the transcription start site (TSS) to the TSS] and down20K [from the transcription end site (TES) to 20 kb downstream], and the number of peaks located in each genomic region was calculated.

Peaks that overlapped by > 50% of the peak length in both lineages (FOXP3-maintaining and FOXP3-losing Treg cells) were defined as ‘common’, and all others were classified as ‘lineage-specific’. A peak was considered associated with a particular gene when at least 1 bp overlapped between the peak location and the gene sequence ranging from 20 kb upstream of the TSS to 20 kb downstream of the TES.

Histone modification profiling by tag density

For each gene, the uniquely mapped reads were summed within 125 bp windows along the gene sequence ranging from 5 kb upstream to the TSS and from the TES to 5 kb downstream. For analysis of gene bodies, the sequence was divided into 100 windows and the reads were summed for each window and normalized to the total read number of the given library and to the total number of bases in the windows.

To study the correlation of the H3K4me3 and H3K27me3 modifications with differential gene expressions, the 1000 most highly expressed and 1000 most repressed (silent) genes were selected from the DGE data. The uniquely mapped reads for each gene set were summed as described above and normalized to the total read number of the two sets and to the total number of bases in the windows.

Quantitative RT-PCR

Total RNA was extracted with the TriPure Reagent (Roche Diagnostic). Reverse transcription was performed using a PrimeScript RT reagent Kit (Takara, Dalian, China). Quantitative real-time PCR was performed with a SYBR Green Kit (Takara) with the primers described in the Supplementary material, Table S1. The CFX-96 real-time PCR system (Bio-Rad, Hercules, CA) was used for all reactions and detection. All expression values were normalized to GAPDH.

Statistical analysis

Mean and SD values were calculated with GraphPad Prism 6 (GraphPad Software, San Diego, CA). Two-tailed unpaired t-test was used in the analysis of mRNA expression.

Results

Human Treg cells lose FOXP3 expression during in vitro expansion

Recent studies of human Treg cells have demonstrated that depleting CD127+ T cells from the CD4+ CD25high population facilitates the isolation of highly enriched FOXP3+ cells with all the functional and molecular characteristics of Treg cells and eliminates the presence of contaminating activated Tconv cells.44–46 Therefore, we isolated CD4+ CD25high CD127low/− T cells from peripheral blood mononuclear cells of healthy volunteers by FACS. The sorted cells showed > 98% purity with regard to the applied sorting gates, and intracellular staining confirmed a high frequency of FOXP3+ cells (96·0 ± 1·8%; Fig. 1a).

Six weeks of in vitro culturing of the sorted Treg cells, using the Treg cell expansion protocol based on stimulation with anti-CD3/anti-CD28 microbeads in the presence of high-dose IL-2, resulted in a profound proliferation (up to 1000-fold expansion; Fig. 1b). The vast majority of cultured cells remained viable after in vitro expansion, although a small fraction was dead presumably for repetitive TCR stimulation and long-term culture (see Supplementary material, Fig. S2). Weekly monitoring of the percentage of FOXP3+ cells by flow cytometry indicated that the human Treg cells developed into a heterogeneous population of FOXP3-maintaining and FOXP3-losing cells upon in vitro expansion, consistent with previous findings.17 The decreased frequency in FOXP3+ cells within the expanded Treg cells was noticeable after 2 weeks, and became more pronounced as the period of in vitro culturing extended and after the second cycle of stimulation (Fig. 1c,d). The average percentage of FOXP3+ cells had fallen to approximately 20% (26·8 ± 11·7%) after 6 weeks of culture. Hence, human Treg cells are not an overwhelmingly stable population and can lose FOXP3 expression during in vitro expansion.

FOXP3-losing Treg cells exhibit a Th gene expression signature

To explore the transcriptional reprogramming of human Treg cells after spontaneous loss of FOXP3 expression, we compared DGE transcriptome data of the FOXP3-maintaining Treg cells to that of the corresponding FOXP3-losing cells. First, the in vitro expanded cells were sorted into FOXP3+ and FOXP3− subpopulations (Fig. 2), and total RNA was extracted for DGE assay. The sequencing tags obtained from each sample were mapped to the reference genome (results are shown in Supplementary material, Table S2). Using stringent filter criteria, we identified a total of 1843 differentially expressed genes in the FOXP3-losing cells (versus FOXP3-maintaining cells). Among these, 1034 genes were up-regulated and 809 genes were down-regulated (Fig. 3a and see Supplementary material, Table S3).

Figure 2.

Figure 2

Open in a new tab

Isolation of FOXP3-maintaining and FOXP3-losing cells from the expanded human regulatory T (Treg) cell population. Treg cells were expanded in vitro for 5 weeks and sorted into FOXP3+ and FOXP3− subpopulations after staining for CD4, CD25 and FOXP3. The sorting gates (left) and the re-analysis of FOXP3+ (upper right) and FOXP3− (lower right) subpopulations are shown.

Figure 3.

Figure 3

Open in a new tab

Transcriptome analysis of expanded FOXP3-maintaining and FOXP3-losing regulatory T (Treg) cells. (a) Expression plots showing differences in gene expression between FOXP3-maintaining and FOXP3-losing cells. Red and green plotted values represent up-regulated and down-regulated genes, respectively, in the FOXP3-losing cells. (b) Heatmaps (pseudocolour scale indicated) of relative gene expression levels for selected gene classes associated with Treg, helper type 1 (Th1), Th2 or Th17 differentiation. *Genes with significant expression differences (fold-change of at least two and P-value of ≤ 0·001) between FOXP3-maintaining and FOXP3-losing subpopulations. (c–e) Gene ontology clustering of differentially expressed genes. (f) Kyoto Encyclopedia of Genes and Genomes pathway analysis for differentially expressed genes in the FOXP3-losing subset.

As expected, the FOXP3-maintaining subpopulation maintained the Treg gene expression profile, including high expression levels of FOXP3, CTLA4, ICOS, IKZF2, LRRC32, SOCS2, SELL and LGALS3, and low expression levels of PDE3B, PDE7A and GPD2. In contrast, the FOXP3-losing cells lost the Treg gene expression signature (Fig. 3b and see Supplementary material, Table S3). Surprisingly, the FOXP3-losing cells showed a remarkable up-regulation of a set of Th2 lineage-associated genes, such as GATA3, GFI1 and IL13. In contrast, the genes TBX21 and RORC, which encode key transcription factors for Th1 and Th17 lineages, respectively, were expressed at low or undetectable levels in the FOXP3-losing cells, although the STAT4 and TNF genes, which are also associated with Th1 and Th17 differentiation, were up-regulated. There was no difference detected between the two subsets for the Th1 or Th17 lineage-associated cytokine genes, such as IFNG, IL6, IL9, IL10, IL17, IL21, IL22 and IL23 (Fig. 3b and see Supplementary material, Table S3). Additionally, the FOXP3-losing cells showed down-regulation of the SOCS2 gene (Fig. 3b), the encoded factor of which inhibits the development of Th2 cells and is required for the stable expression of Foxp3 in iTreg cells.47 Furthermore, quantitative RT-PCR analysis for the selected signature genes showed that the FOXP3-losing cells display a different mRNA expression signature from not only FOXP3-maintaining Treg cells but also expanded CD4+ Tconv cells (see Supplementary material, Fig. S3). Hence, consistent with the recently reported findings by Hansmann et al.,48 we found that human Treg cells can convert into cells with a gene expression signature dominated by the Th2 lineage-associated genes, upon spontaneous loss of FOXP3 expression.

FOXP3 and GATA3 are the master regulators for Treg and Th2 cell lineages, respectively. So, we assessed the conversion dynamics of human Treg cells during in vitro culture by analysing expression levels of the FOXP3 and GATA3 genes in expanded Treg cells at different time-points. The results showed that the FOXP3 expression level reduced gradually as the period of in vitro culturing extended (Fig. 4a), which is in line with the results from flow cytometry analysis (Fig. 1d). In contrast, the GATA3 expression level increased gradually as the expansion time extended (Fig. 4b). Hence, the conversion of human Treg cells is a gradual process, which depends on spontaneous loss of FOXP3 expression.

Figure 4.

Figure 4

Open in a new tab

Dynamic analysis of FOXP3 and GATA3 expression levels in expanded regulatory T (Treg) cells. The expression levels of gene FOXP3 (a) and GATA3 (b) in expanded Treg cells at different time-points. Bars represent means ± SD.

To investigate the functional features of the molecular reprogramming induced by spontaneous loss of FOXP3, we conducted gene ontology and KEGG pathway analysis. Gene ontology analysis revealed a substantial enrichment for genes encoding cytoplasmic proteins with involvement in immunity, leucocyte activation, lymphocyte activation and cell activation (Fig. 3c–e). KEGG analysis indicated a strong association of the differentially expressed genes with rheumatoid arthritis, the TCR signalling pathway, type 1 diabetes mellitus and the intestinal immune network for IgA production (Fig. 3f). Collectively, these results suggest that human Treg cells can be converted into cells expressing the Th cell transcriptional programme upon loss of the lineage-associated transcription factor FOXP3.

Genome-wide H3K4me3 and H3K27me3 profiles of FOXP3-maintaining and FOXP3-losing cells

To reveal the epigenetic features present during the process of Treg FOXP3 loss in vitro, global maps of H3K4me3 and H3K27me3 modifications were generated via the ChIP-Seq approach (results shown in Supplementary material, Table S4). In total, 24 410 H3K4me3 and 1334 H3K27me3 peaks were found in the FOXP3-maintaining cells, while 21 762 H3K4me3 and 336 H3K27me3 peaks were found in the FOXP3-losing cells.

The FOXP3-maintaining and FOXP3-losing cell subsets displayed an appreciable percentage of comparable distribution for both H3K4me3 and H3K27me3 peaks in each of the five genomic regions assessed. In particular, the H3K4me3 peaks were enriched in regions of intergenic (˜ 69%), intron (˜ 65%), up20K (˜ 56%) and exon (˜ 48%), and the H3K27me3 peaks were enriched in intergenic regions (74%; Fig. 5a). However, when compared with the amount of H3K4me3 peaks in each genomic region, it was found that the amounts of H3K27me3 peaks were significantly lower in the regions of up20K (˜ 14% for both cell subsets), intron (˜ 18% and ˜ 15% for FOXP3-maintaining and FOXP3-losing cells, respectively), and down20K (˜ 9% and 11% for FOXP3-maintaining and FOXP3-losing cells, respectively); moreover, the exon regions had only 2% peaks in the FOXP3-maintaining cells and 6% peaks in FOXP3-losing cells (Fig. 5a). Examination of H3K4me3 reads located within gene bodies and within their 5′- and 3′-5-kb extended regions revealed an enrichment of H3K4me3 marks near the TSS (Fig. 5b). Notably, the H3K27me3 levels were high in the region encompassing 5 kb upstream to the TSS, but were remarkably lower within the gene bodies (Fig. 5b). Hence, the H3K4me3 modifications in human FOXP3-maintaining and FOXP3-losing cells were most prominently associated with gene promoters, whereas the majority of the H3K27me3 peaks were associated with intergenic regions, consistent with previous reports of data from other relevant studies.31,49–51

Figure 5.

Figure 5

Open in a new tab

Genome-wide maps of H3K4me3 and H3K27me3 modifications in FOXP3-maintaining and FOXP3-losing cells after regulatory T (Treg) cell in vitro expansion. (a) Distribution of H3K4me3 (left) and H3K27me3 (right) peaks among different genomic regions. (b) Coverage depth of H3K4me3 (left) and H3K27me3 (right) reads among the genomic regions. For each gene, uniquely mapped tags were summed in windows equal to 1% of the gene-body length or in 125-bp windows for the regions encompassing 5 kb upstream to the TSS or the TES to 5 kb downstream, respectively. (c) Lineage-specific H3K4me3 (left) and H3K27me3 (right) peaks are shown. Common peaks showed overlap (> 50% length of the small peak) between the FOXP3-maintaining cells and the FOXP3-losing cells. Specific peaks show no overlap. (d) Venn diagram showing the differences in genes related to H3K4me3 (left) or H3K27me3 (right) modifications.

Comparison of the common and lineage-specific H3K4me3 and H3K27me3 peaks in the FOXP3-maintaining and FOXP3-losing cells revealed that 22% of the H3K4me3 peaks and 82% of the H3K27me3 peaks were unique to the FOXP3-maintaining Treg cells (Fig. 5c and see Supplementary material, Table S5). Interestingly, only 12% of the H3K4me3 peaks and 47% of the H3K27me3 peaks were unique to the FOXP3-losing cells, and this epigenetic feature was similar to that of Th cells observed in a previous study.49 When the particular genes associated with the H3K4me3 or H3K27me3 peaks of the two cell subsets were compared, more unique genes were found in the FOXP3-maintaining cells than in the FOXP3-losing cells (1761 versus 456 for H3K4me3; 778 versus 180 for H3K27me3; Fig. 5d). Additionally, the majority of the genes associated with H3K4me3 were common between the two cell subsets, which contrasts with the fact that only a small number of genes related to H3K27me3 peaks were common (Fig. 5d). Taken together, these results indicate that the FOXP3-losing cells display a global epigenetic modification pattern that is distinct from that of the FOXP3-maintaining cells.

Correlation of H3K4me3 and H3K27me3 modifications with genes’ expression

To reveal the functional consequences of H3K4me3 and H3K27me3 on associated genes, the data of ChIP-Seq and DGE (2000 selected genes, as described) were comparatively analysed. As expected, the initial overview of the histone modification profiles in relation to genes showed that H3K4me3 levels were elevated surrounding the TSSs and peaked at the TSSs, whereas H3K27me3 levels were elevated in the region from 2 kb upstream to the TSSs and were dramatically lower at the TSSs. When the genes’ active state was considered, higher levels of H3K4me3 surrounding the TSSs were found for active genes than for silent genes (Fig. 6a); in addition, the levels of H3K27me3 across the gene bodies and 5′- and 3′-5-kb extended regions were higher for silent genes than for active genes (Fig. 6b). These results indicate that H3K4me3, as a permissive mark, is correlated with gene activation, whereas H3K27me3, as a repressive mark, is correlated with gene silencing, consistent with previous reports.30,32

Figure 6.

Figure 6

Open in a new tab

H3K4me3 and H3K27me3 profiles indicate the gene active state. The normalized tag numbers of H3K4me3 (a) or H3K27me3 (b) for 1000 highly active or 1000 silent genes across the gene bodies (x-axis).

Plasticity of histone methylation affects Treg cell signature genes’ expression

To investigate the role of epigenetic modification patterns in the determination of the fate of Treg cells, the permissive and repressive histone modifications related to Treg signature genes were compared between the FOXP3-maintaining and FOXP3-losing cells. In agreement with previous findings for murine Treg cells,49 the human cell data in the current study indicated that the promoter region of the FOXP3 gene had substantial H3K4me3 modification but almost no H3K27me3 modification. As expected, the robust H3K4me3 marks of this promoter region were absent in the FOXP3-losing cells, corresponding to its down-regulated expression. However, the FOXP3-losing cells did not show a corresponding increase in the amount of H3K27me3 in this gene locus (Fig. 7).

Figure 7.

Figure 7

Open in a new tab

Expression levels and epigenetic modification patterns of regulatory T (Treg) cell signature genes. (a) The mRNA expression levels of Treg signature genes, as determined by digital gene expression analysis. (b) The distribution patterns of H3K4me3 (left) and H3K27me3 (right) surrounding the FOXP3, CTLA4 and LRRC32 genes.

The H3K4me3 and H3K27me3 patterns were also examined for other Treg lineage-specific genes, such as CTLA4, LRRC32, LGALS3 and SELL, all of which were found to be down-regulated in the FOXP3-losing cells. All of these genes showed high amounts of H3K4me3 in the FOXP3-maintaining cells, but reduced or nearly undetectable amounts in the FOXP3-losing cells; these results agree with the corresponding gene expressions detected by DGE analysis (Fig. 7 and see Supplementary material, Fig. S4). However, their promoter regions and gene bodies all possessed low amounts of H3K27me3 in both of the expanded cell subsets and there were no significant differences between the two subsets (Fig. 7b). Hence, the lineage-specific gene expression levels appear to correlate with the amounts of H3K4me3, but not necessarily with the amounts of H3K27me3.

As shown by DGE analysis, the FOXP3-losing cells showed a marked up-regulation of a set of genes related to the Th2 lineage (Fig. 8a). When the histone modification patterns for these genes were compared between the FOXP3-maintaining cells and the FOXP3-losing cells correlations were found with the differential gene expression patterns. For example, the IL4 and IL5 genes, which encode Th2 lineage-associated cytokines, showed higher amounts of H3K4me3 in the FOXP3-losing cells (Fig. 8). However, both cell subsets showed remarkably robust levels of H3K4me3 in the promoter of the GATA3, consistent with a previous finding in mice showing that the Gata3 promoter was marked by H3K4me3 in non-expressing lineages, as well as in Th2 cells.49 Analysis of the H3K27me3 levels in these Th2 lineage-associated genes showed a similar low trend for both cell subsets (Fig. 8), as was observed in the Treg signature genes’ loci. Hence, H3K4me3 modification may play an essential role in Treg cell fate determination by affecting signature genes’ expression; but the role of H3K27me3 modification appears to be more complex and additional studies are needed.

Figure 8.

Figure 8

Open in a new tab

Expression levels and epigenetic modification patterns of T helper type 2 (Th2) lineage-associated genes. (a) The mRNA expression levels of Th2 lineage-associated genes, as determined by digital gene expression analysis. (b) The distribution patterns of H3K4me3 (left) and H3K27me3 (right) surrounding the IL4, IL5 and GATA3 genes.

Discussion

Accumulating evidence is revealing the unstable nature of the Treg cell lineage both in vitro and in vivo, with a subset of this cell population losing FOXP3 expression under certain conditions,17,19–24 piquing interest in the fate of Treg cells after FOXP3 loss and the underlying mechanism of Treg cell plasticity. The study described herein was designed to investigate these issues by analysing the gene expression and histone methylation profiles of FOXP3-maintaining and FOXP3-losing cell subsets generated upon Treg cell in vitro expansion. The data indicated that the FOXP3-losing Treg cells were reprogrammed into cells with a gene expression signature dominated by Th2 lineage-associated genes and that histone methylation may contribute to this reprogramming.

CD45RA− Treg cells were recently demonstrated in another study to convert into effector Th2 cells after expansion in vitro.48 Although that previous study identified the CD45RA− memory-type Treg cells as the main source of the FOXP3-losing cells in culture, it is also known that CD45RA+ Treg cells, widely considered as an optimal candidate for in vitro expansion, can lose FOXP3 expression during in vitro expansion, especially after longer culture periods and in conditions of repetitive TCR stimulation.16,17 Therefore, in the present study, we selected the bulk Treg cell population, containing both CD45RA+ and CD45RA− cells, as the starting population to explore the fate of human Treg cells during in vitro expansion. To avoid contamination of isolated human Treg cells with activated Tconv cells, CD127high cells were depleted from the CD4+ CD25high population, as CD127 can be used to distinguish Treg cells from activated CD25+ Tconv cells. However, even highly purified FOXP3+ Treg cells can readily lose FOXP3 expression in culture; with this in mind, we monitored the consistent negative or low expression of CD127 during our study's in vitro expansion (data not shown) to help ensure the absence of any significant contamination with Tconv cells and so-called ‘induced’ Treg cells that only transiently express FOXP3 and can revert to Tconv cells during in vitro culture. When our results were considered together with those from the previous CD45RA+ Treg cell subpopulation study,48 it can be concluded that a subset of human Treg cells can lose the Treg gene expression signature after loss of FOXP3 expression, acquiring a Th gene expression signature that is dominated by Th2 lineage-associated genes, no matter which subpopulation this subset was derived from.

Accumulating studies are revealing a close relationship between Treg cells and Th2 cells.52 Previous studies in mice have shown that decreased expression of Foxp3 results in the conversion of the Treg cells into effector T cells, especially Th2 cells.13,14 The data presented herein demonstrate that human Treg cells can be reprogrammed into cells expressing Th2 lineage-associated genes upon spontaneous loss of FOXP3 expression, further supporting the theory of plasticity between Treg and Th2 cells.52 Such plasticity may be illustrated by the reciprocal regulation and the functional interactions of FOXP3 and GATA3, critical transcription factors for Treg and Th2 cell lineages, respectively;53 indeed, it has been shown that GATA3 becomes up-regulated in a portion of Treg cells upon in vitro TCR stimulation in the presence of IL-2.53 Recent studies have also demonstrated that the transcription factor function of GATA3 is essential for Foxp3 expression and Treg cell function, through directly binding to a conserved element in the Foxp3 locus together with Foxp3 itself.54,55 Furthermore, enforced expression of GATA3 was shown to limit Treg cell conversion toward the effector T-cell phenotype in inflammatory settings,54 whereas another study showed GATA3 acting as a potent inhibitor of TGF-β-induced Foxp3+ iTreg cell generation from naive CD4+ T cells.56,57 These apparently discordant findings may be reconciled by a recent study of the GATA3 and Foxp3 functional interactions; the results indicated that these factors, either in a cooperative fashion or in an antagonistic manner, cooperate to regulate their own expression and the downstream Foxp3-dependent transcriptional programme.53 Therefore, we speculate that spontaneous loss of FOXP3 and up-regulation of GATA3 in human Treg cells during expansion in vitro may result from perturbation of the complicated relationship between FOXP3 and GATA3.

Besides transcription factors, epigenetic modifications, such as methylation and acetylation, are important regulators of gene expression. Previous comparative analysis of histone modification status, as was done in the present study, has provided insights into the plasticity of CD4+ T-cell subsets.49 The present study revealed that loss of FOXP3 expression from Treg cells was associated with a differential profile of H3K4me3 modification, with abundant H3K4me3 marks detected at the FOXP3 locus in the FOXP3-maintaining cells but not in the FOXP3-losing cells. This finding is consistent with results from a previous murine study, which showed that the Foxp3 proximal promoter region was marked by H3K4me3 modification only in Treg cells and not in other CD4+ T-cell lineages.49 In contrast, abundant H3K4me3 marks were detected at the GATA3 locus in both FOXP3-maintaining and FOXP3-losing cells in the present study. In fact, the Gata3 locus was found to be modified by H3K4me3 in all CD4+ T-cell subsets previously studied, including Foxp3+ Tregs, and not only in Th2 cells,49 which indicated a potential plasticity for Gata3 expression. Here, we demonstrated a similar plasticity feature and found that up-regulation of GATA3 did occur in FOXP3-losing cells. We therefore speculate that the abundant permissive H3K4me3 marks just provide a prerequisite for the up-regulation of GATA3.

Epigenetic inheritance during cell division is known to be crucial for maintaining differential gene expression patterns in cell lineages.58 At present, the extrinsic and intrinsic signals that regulate epigenetic marks at the FOXP3 locus from one cell cycle to another remain unknown. The relationship of histone modifications to Foxp3 expression was addressed in a previous study of murine naive CD4+ T cells,59 and the findings indicated that H3K4me2 and H3K4me3 modifications could be induced at the Foxp3 promoter and intronic enhancer after 18 hr of TCR stimulation. These regions were also found to be modified by H3K4me2 and H3K4me3 in Treg cells, and these epigenetic marks were lost upon continuous TCR signalling in the naive CD4+ T cells, resulting in subsequent inhibition of the Foxp3 expression.59 These findings suggest that TCR stimulation can lead to changes in epigenetic marks at the FOXP3 locus. Therefore, we cannot exclude the possibility that the observed reduction of H3K4me3 marks at the FOXP3 locus in Treg cells of the present study was due to repetitive TCR stimulation or other extrinsic signals present during the long-term in vitro culture.

Of course, other epigenetic marks may also be implicated in the course of Treg FOXP3 loss, such as CpG DNA methylation and histone deacetylation.28 Furthermore, it was shown that inhibition of DNA demethylation through treatment with the DNA methyltransferase-1 inhibitor 5-aza-2-deoxycytidine (5-Aza) can induce stable FOXP3/Foxp3 expression in conventional T cells.60,61 Hence, epigenetic intervention, such as administration of DNA methyltransferase and histone/protein deacetylase inhibitors, may be an optional approach to maintain Treg cell stability. Global genome-wide analysis of epigenetic marks in Treg cells and their subpopulations will enrich our capability to design more preferable epigenetic therapy. Additionally, other strategies to stabilize FOXP3 expression and enhance Treg cell generation – for example, through treatment with cytokines or by interfering with intracellular signalling pathways or microRNAs – should also be explored in the future studies. These investigations will provide a better approach to generate functionally stable Treg cells in vitro for safe and effective in vivo usage.

In summary, the comparative transcriptome analyses described herein identified and characterized the conversion of human Treg cells into cells with Th cell programmes, especially Th2, upon loss of FOXP3 expression. Furthermore, these analyses provide the first map of the histone modifications that accompany the conversion of human Treg cells into Th-like cells, and indicate that histone modifications may contribute to Treg cell reprogramming. Ultimately, these findings expand our understanding of Treg cell plasticity and its epigenetic mechanisms in humans.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant No. 81202334, 30901479 and 31200668).

Author contributions

YL, BN and YZ designed the study. HH, YT, ZT and YC performed the experiments. YZ, YT and ZL. XY analysed the data. HH, BN and YL wrote the paper.

Disclosures

The authors declare no conflict of interest.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Figure S1. Sorting gates (left) and re-analysis (right) of human CD4+ CD25high CD127low/− and CD4+ CD25− CD127+T-cell subpopulations.

Figure S2. Live-cell analysis for in vitro cultured human T regulatory cells.

Figure S3. FOXP3-losing cells exhibit different mRNA expression from FOXP3-maintaining cells and CD4+ conventional T cells.

Figure S4. Expression levels and epigenetic modification patterns of regulatory T cell signature genes SELL and LGALS3.

Table S1. PCR primers used for quantitative PCR in Figure 4 and Figure S3.

Table S2. The mapping results of digital gene expression tags.

Table S3. Differentially expressed gene list.

Table S4. The mapping results of H3K4me3 and H3K27me3 reads.

Table S5. The common and lineage-specific peaks.

imm0141-0362-sd1.pdf (1.5MB, pdf)

References

  • 1.Sakaguchi S, Miyara M, Costantino CM, Hafler DA. FOXP3+ regulatory T cells in the human immune system. Nat Rev Immunol. 2010;10:490–500. doi: 10.1038/nri2785. [DOI] [PubMed] [Google Scholar]
  • 2.Tang Q, Henriksen KJ, Bi M, et al. In vitro-expanded antigen-specific regulatory T cells suppress autoimmune diabetes. J Exp Med. 2004;199:1455–65. doi: 10.1084/jem.20040139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Mottet C, Uhlig HH, Powrie F. Cutting edge: cure of colitis by CD4+CD25+ regulatory T cells. J Immunol. 2003;170:3939–43. doi: 10.4049/jimmunol.170.8.3939. [DOI] [PubMed] [Google Scholar]
  • 4.Morgan ME, Flierman R, Duivenvoorde van LM, Witteveen HJ, Ewijk van W, Laar van JM, Vries de RR, Toes RE. Effective treatment of collagen-induced arthritis by adoptive transfer of CD25+ regulatory T cells. Arthritis Rheum. 2005;52:2212–21. doi: 10.1002/art.21195. [DOI] [PubMed] [Google Scholar]
  • 5.Trzonkowski P, Bieniaszewska M, Juscinska J, Dobyszuk A, Krzystyniak A, Marek N, Mysliwska J, Hellmann A. First-in-man clinical results of the treatment of patients with graft versus host disease with human ex vivo expanded CD4+CD25+CD127- T regulatory cells. Clin Immunol. 2009;133:22–6. doi: 10.1016/j.clim.2009.06.001. [DOI] [PubMed] [Google Scholar]
  • 6.Brunstein CG, Miller JS, Cao Q, et al. Infusion of ex vivo expanded T regulatory cells in adults transplanted with umbilical cord blood: safety profile and detection kinetics. Blood. 2011;117:1061–70. doi: 10.1182/blood-2010-07-293795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ianni Di M, Falzetti F, Carotti A, et al. Tregs prevent GVHD and promote immune reconstitution in HLA-haploidentical transplantation. Blood. 2011;117:3921–8. doi: 10.1182/blood-2010-10-311894. [DOI] [PubMed] [Google Scholar]
  • 8.Abbas AK, Benoist C, Bluestone JA, et al. Regulatory T cells: recommendations to simplify the nomenclature. Nat Immunol. 2013;14:307–8. doi: 10.1038/ni.2554. [DOI] [PubMed] [Google Scholar]
  • 9.Curotto de Lafaille MA, Lafaille JJ. Natural and adaptive foxp3+ regulatory T cells: more of the same or a division of labor? Immunity. 2009;30:626–35. doi: 10.1016/j.immuni.2009.05.002. [DOI] [PubMed] [Google Scholar]
  • 10.Zheng Y, Josefowicz SZ, Kas A, Chu TT, Gavin MA, Rudensky AY. Genome-wide analysis of Foxp3 target genes in developing and mature regulatory T cells. Nature. 2007;445:936–40. doi: 10.1038/nature05563. [DOI] [PubMed] [Google Scholar]
  • 11.Marson A, Kretschmer K, Frampton GM, et al. Foxp3 occupancy and regulation of key target genes during T-cell stimulation. Nature. 2007;445:931–5. doi: 10.1038/nature05478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Zheng Y, Rudensky AY. Foxp3 in control of the regulatory T cell lineage. Nat Immunol. 2007;8:457–62. doi: 10.1038/ni1455. [DOI] [PubMed] [Google Scholar]
  • 13.Williams LM, Rudensky AY. Maintenance of the Foxp3-dependent developmental program in mature regulatory T cells requires continued expression of Foxp3. Nat Immunol. 2007;8:277–84. doi: 10.1038/ni1437. [DOI] [PubMed] [Google Scholar]
  • 14.Wan YY, Flavell RA. Regulatory T-cell functions are subverted and converted owing to attenuated Foxp3 expression. Nature. 2007;445:766–70. doi: 10.1038/nature05479. [DOI] [PubMed] [Google Scholar]
  • 15.Miyara M, Yoshioka Y, Kitoh A, et al. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity. 2009;30:899–911. doi: 10.1016/j.immuni.2009.03.019. [DOI] [PubMed] [Google Scholar]
  • 16.Hoffmann P, Eder R, Boeld TJ, Doser K, Piseshka B, Andreesen R, Edinger M. Only the CD45RA+ subpopulation of CD4+CD25high T cells gives rise to homogeneous regulatory T-cell lines upon in vitro expansion. Blood. 2006;108:4260–7. doi: 10.1182/blood-2006-06-027409. [DOI] [PubMed] [Google Scholar]
  • 17.Hoffmann P, Boeld TJ, Eder R, et al. Loss of FOXP3 expression in natural human CD4+CD25+ regulatory T cells upon repetitive in vitro stimulation. Eur J Immunol. 2009;39:1088–97. doi: 10.1002/eji.200838904. [DOI] [PubMed] [Google Scholar]
  • 18.d'Hennezel E, Yurchenko E, Sgouroudis E, Hay V, Piccirillo CA. Single-cell analysis of the human T regulatory population uncovers functional heterogeneity and instability within FOXP3+ cells. J Immunol. 2011;186:6788–97. doi: 10.4049/jimmunol.1100269. [DOI] [PubMed] [Google Scholar]
  • 19.Xu L, Kitani A, Fuss I, Strober W. Cutting edge: regulatory T cells induce CD4+CD25-Foxp3- T cells or are self-induced to become Th17 cells in the absence of exogenous TGF-beta. J Immunol. 2007;178:6725–9. doi: 10.4049/jimmunol.178.11.6725. [DOI] [PubMed] [Google Scholar]
  • 20.Yang XO, Nurieva R, Martinez GJ, et al. Molecular antagonism and plasticity of regulatory and inflammatory T cell programs. Immunity. 2008;29:44–56. doi: 10.1016/j.immuni.2008.05.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Koenen HJ, Smeets RL, Vink PM, Rijssen van E, Boots AM, Joosten I. Human CD25highFoxp3pos regulatory T cells differentiate into IL-17-producing cells. Blood. 2008;112:2340–52. doi: 10.1182/blood-2008-01-133967. [DOI] [PubMed] [Google Scholar]
  • 22.Duarte JH, Zelenay S, Bergman ML, Martins AC, Demengeot J. Natural Treg cells spontaneously differentiate into pathogenic helper cells in lymphopenic conditions. Eur J Immunol. 2009;39:948–55. doi: 10.1002/eji.200839196. [DOI] [PubMed] [Google Scholar]
  • 23.Komatsu N, Mariotti-Ferrandiz ME, Wang Y, Malissen B, Waldmann H, Hori S. Heterogeneity of natural Foxp3+ T cells: a committed regulatory T-cell lineage and an uncommitted minor population retaining plasticity. Proc Natl Acad Sci USA. 2009;106:1903–8. doi: 10.1073/pnas.0811556106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhou X, Bailey-Bucktrout SL, Jeker LT, et al. Instability of the transcription factor Foxp3 leads to the generation of pathogenic memory T cells in vivo. Nat Immunol. 2009;10:1000–7. doi: 10.1038/ni.1774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Baron U, Floess S, Wieczorek G, et al. DNA demethylation in the human FOXP3 locus discriminates regulatory T cells from activated FOXP3(+) conventional T cells. Eur J Immunol. 2007;37:2378–89. doi: 10.1002/eji.200737594. [DOI] [PubMed] [Google Scholar]
  • 26.Floess S, Freyer J, Siewert C, et al. Epigenetic control of the foxp3 locus in regulatory T cells. PLoS Biol. 2007;5:e38. doi: 10.1371/journal.pbio.0050038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kim HP, Leonard WJ. CREB/ATF-dependent T cell receptor-induced FoxP3 gene expression: a role for DNA methylation. J Exp Med. 2007;204:1543–51. doi: 10.1084/jem.20070109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lal G, Bromberg JS. Epigenetic mechanisms of regulation of Foxp3 expression. Blood. 2009;114:3727–35. doi: 10.1182/blood-2009-05-219584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Haiqi H, Yong Z, Yi L. Transcriptional regulation of Foxp3 in regulatory T cells. Immunobiology. 2011;216:678–85. doi: 10.1016/j.imbio.2010.11.002. [DOI] [PubMed] [Google Scholar]
  • 30.Roh TY, Cuddapah S, Cui K, Zhao K. The genomic landscape of histone modifications in human T cells. Proc Natl Acad Sci USA. 2006;103:15782–7. doi: 10.1073/pnas.0607617103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Barski A, Cuddapah S, Cui K, et al. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–37. doi: 10.1016/j.cell.2007.05.009. [DOI] [PubMed] [Google Scholar]
  • 32.Wang Z, Zang C, Rosenfeld JA, et al. Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet. 2008;40:897–903. doi: 10.1038/ng.154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Tao R, Zoeten de EF, Ozkaynak E, et al. Deacetylase inhibition promotes the generation and function of regulatory T cells. Nat Med. 2007;13:1299–307. doi: 10.1038/nm1652. [DOI] [PubMed] [Google Scholar]
  • 34.Lucas JL, Mirshahpanah P, Haas-Stapleton E, Asadullah K, Zollner TM, Numerof RP. Induction of Foxp3+ regulatory T cells with histone deacetylase inhibitors. Cell Immunol. 2009;257:97–104. doi: 10.1016/j.cellimm.2009.03.004. [DOI] [PubMed] [Google Scholar]
  • 35.t Hoen PA, Ariyurek Y, Thygesen HH, et al. Deep sequencing-based expression analysis shows major advances in robustness, resolution and inter-lab portability over five microarray platforms. Nucleic Acids Res. 2008;36:e141. doi: 10.1093/nar/gkn705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Morrissy AS, Morin RD, Delaney A, et al. Next-generation tag sequencing for cancer gene expression profiling. Genome Res. 2009;19:1825–35. doi: 10.1101/gr.094482.109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res. 1997;7:986–95. doi: 10.1101/gr.7.10.986. [DOI] [PubMed] [Google Scholar]
  • 38.Kanehisa M, Araki M, Goto S, et al. KEGG for linking genomes to life and the environment. Nucleic Acids Res. 2008;36:D480–4. doi: 10.1093/nar/gkm882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lee TI, Johnstone SE, Young RA. Chromatin immunoprecipitation and microarray-based analysis of protein location. Nat Protoc. 2006;1:729–48. doi: 10.1038/nprot.2006.98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Tian Y, Jia Z, Wang J, et al. Global Mapping of H3K4me1 and H3K4me3 Reveals the Chromatin State-Based Cell Type-Specific Gene Regulation in Human Treg Cells. PLoS ONE. 2011;6:e27770. doi: 10.1371/journal.pone.0027770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Li R, Li Y, Kristiansen K, Wang J. SOAP: short oligonucleotide alignment program. Bioinformatics. 2008;24:713–4. doi: 10.1093/bioinformatics/btn025. [DOI] [PubMed] [Google Scholar]
  • 42.Li R, Yu C, Li Y, Lam TW, Yiu SM, Kristiansen K, Wang J. SOAP2: an improved ultrafast tool for short read alignment. Bioinformatics. 2009;25:1966–7. doi: 10.1093/bioinformatics/btp336. [DOI] [PubMed] [Google Scholar]
  • 43.Zhang Y, Liu T, Meyer CA, et al. Model-based analysis of ChIP-Seq (MACS) Genome Biol. 2008;9:R137. doi: 10.1186/gb-2008-9-9-r137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Liu W, Putnam AL, Xu-Yu Z, et al. CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells. J Exp Med. 2006;203:1701–11. doi: 10.1084/jem.20060772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Seddiki N, Santner-Nanan B, Martinson J, et al. Expression of interleukin (IL)-2 and IL-7 receptors discriminates between human regulatory and activated T cells. J Exp Med. 2006;203:1693–700. doi: 10.1084/jem.20060468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Hartigan-O'Connor DJ, Poon C, Sinclair E, McCune JM. Human CD4+ regulatory T cells express lower levels of the IL-7 receptor alpha chain (CD127), allowing consistent identification and sorting of live cells. J Immunol Methods. 2007;319:41–52. doi: 10.1016/j.jim.2006.10.008. [DOI] [PubMed] [Google Scholar]
  • 47.Knosp CA, Schiering C, Spence S, et al. Regulation of Foxp3+ inducible regulatory T cell stability by SOCS2. J Immunol. 2013;190:3235–45. doi: 10.4049/jimmunol.1201396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Hansmann L, Schmidl C, Kett J, Steger L, Andreesen R, Hoffmann P, Rehli M, Edinger M. Dominant Th2 differentiation of human regulatory T cells upon loss of FOXP3 expression. J Immunol. 2012;188:1275–82. doi: 10.4049/jimmunol.1102288. [DOI] [PubMed] [Google Scholar]
  • 49.Wei G, Wei L, Zhu J, et al. Global mapping of H3K4me3 and H3K27me3 reveals specificity and plasticity in lineage fate determination of differentiating CD4+ T cells. Immunity. 2009;30:155–67. doi: 10.1016/j.immuni.2008.12.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Pan G, Tian S, Nie J, et al. Whole-genome analysis of histone H3 lysine 4 and lysine 27 methylation in human embryonic stem cells. Cell Stem Cell. 2007;1:299–312. doi: 10.1016/j.stem.2007.08.003. [DOI] [PubMed] [Google Scholar]
  • 51.Mikkelsen TS, Ku M, Jaffe DB, et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007;448:553–60. doi: 10.1038/nature06008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Chapoval S, Dasgupta P, Dorsey NJ, Keegan AD. Regulation of the T helper cell type 2 (Th2)/T regulatory cell (Treg) balance by IL-4 and STAT6. J Leukoc Biol. 2010;87:1011–8. doi: 10.1189/jlb.1209772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Rudra D, deRoos P, Chaudhry A, et al. Transcription factor Foxp3 and its protein partners form a complex regulatory network. Nat Immunol. 2012;13:1010–9. doi: 10.1038/ni.2402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Wang Y, Su MA, Wan YY. An essential role of the transcription factor GATA-3 for the function of regulatory T cells. Immunity. 2011;35:337–48. doi: 10.1016/j.immuni.2011.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Wohlfert EA, Grainger JR, Bouladoux N, et al. GATA3 controls Foxp3(+) regulatory T cell fate during inflammation in mice. J Clin Invest. 2011;121:4503–15. doi: 10.1172/JCI57456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Mantel PY, Kuipers H, Boyman O, et al. GATA3-driven Th2 responses inhibit TGF-beta1-induced FOXP3 expression and the formation of regulatory T cells. PLoS Biol. 2007;5:e329. doi: 10.1371/journal.pbio.0050329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Wei J, Duramad O, Perng OA, Reiner SL, Liu YJ, Qin FX. Antagonistic nature of T helper 1/2 developmental programs in opposing peripheral induction of Foxp3+ regulatory T cells. Proc Natl Acad Sci USA. 2007;104:18169–74. doi: 10.1073/pnas.0703642104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Probst AV, Dunleavy E, Almouzni G. Epigenetic inheritance during the cell cycle. Nat Rev Mol Cell Biol. 2009;10:192–206. doi: 10.1038/nrm2640. [DOI] [PubMed] [Google Scholar]
  • 59.Sauer S, Bruno L, Hertweck A, et al. T cell receptor signaling controls Foxp3 expression via PI3K, Akt, and mTOR. Proc Natl Acad Sci USA. 2008;105:7797–802. doi: 10.1073/pnas.0800928105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Polansky JK, Kretschmer K, Freyer J, et al. DNA methylation controls Foxp3 gene expression. Eur J Immunol. 2008;38:1654–63. doi: 10.1002/eji.200838105. [DOI] [PubMed] [Google Scholar]
  • 61.Lal G, Zhang N, Touw van der W, et al. Epigenetic regulation of Foxp3 expression in regulatory T cells by DNA methylation. J Immunol. 2009;182:259–73. doi: 10.4049/jimmunol.182.1.259. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Sorting gates (left) and re-analysis (right) of human CD4+ CD25high CD127low/− and CD4+ CD25− CD127+T-cell subpopulations.

Figure S2. Live-cell analysis for in vitro cultured human T regulatory cells.

Figure S3. FOXP3-losing cells exhibit different mRNA expression from FOXP3-maintaining cells and CD4+ conventional T cells.

Figure S4. Expression levels and epigenetic modification patterns of regulatory T cell signature genes SELL and LGALS3.

Table S1. PCR primers used for quantitative PCR in Figure 4 and Figure S3.

Table S2. The mapping results of digital gene expression tags.

Table S3. Differentially expressed gene list.

Table S4. The mapping results of H3K4me3 and H3K27me3 reads.

Table S5. The common and lineage-specific peaks.

imm0141-0362-sd1.pdf (1.5MB, pdf)

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES